Population: 9,407 Search:

The Stuffopolis.net Multiview Forum
Topic Tags: Laptops, ASUS, ASUS-G51Jx-A1

Profile | Register | Search

The Multiview Forum This Topic is 16 Pages Long:    1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Views: 8170   Post Reply 
Review: G51JX-A1 Review

 Posted - 2/3/2010 11:28:51 PM  - Show Profile  Edit Message

Site Administrator
2020 Contributions
Title: G51JX-A1 Review

Item Resource Page: ASUS-G51Jx-A1

Pros: TBD

Cons: TBD

Price Paid: $1539 CAD + tax

Overall Rating: 8/10

G51Jx-A1 Review

My specs:
Model:   G51JX-A1
Motherboard: G60JX
Screen: 15.6" FHD (1920x1080) LCD w/ LED back lighting
Processor: Intel Quad Core i7-720QM (1.60GHz)
Chipset: Intel HM55
RAM: 6GB Kingston DDR3 1333MHz w/4 Slots
Hard Disk: 500G 7200RPM
Video Card: NV GTS 360M 1GB GDDR5 VRAM
Network: 802.11B/G/N, 10/100/1000 Ethernet
OS: Windows 7 Home Premium (64 bit)
Weight: 7.26 lbs (Manufacturer Stated Weight)
ODD: 8x Super-Multi DVDRW Dual layer (TSSTcorp CDDVDW TS-L633C)
Webcam: 2.0M
Ports (USB/FW/eSATA/IR/BT): 4/1/1/1/1

My unit also came with a green/gray ASUS ROG Backpack and a ASUS Gaming Mouse Razer Salmosa (which the retailer did not even mention)

The box has G60X as the the motherboard, and the unit has G51J written on it, so I was a tiny bit concerned that I might not have the right unit, it being the first day they were available and everything - but it is indeed a G51JX :).

Screen / First Design Impressions

My first impression - they could really have fit a 16" Screen with how big the bezel is - it's huge! I like the design though, and the black finish looks good. The system is heavier than I expected, and larger too - but not by too much. The various parts of the system pretty much all seem to be fingerprint magnets.

The screen looks great though, it's very bright and the contrast seems to be ok. I did a test with full screen red, green, blue, white and black images and the LCD is pixel perfect from my close examination - so I don't think I'll need to use the ZBD guarantee. The screen resolution is certainly comfortable for me. The resolution is a bit larger than the 1680x1050 I'm used to on my Z96JS, but the screen is also larger by .2", and with my Xperia X1 phone having a 800x600 resolution a 3.0" screen I really can't see myself enjoying a much smaller resolution on this size panel. Viewing angles are very good on the horizontal and OK on the vertical with a narrower range. The screen assembly is solid and the hinges work well with no flex.

The blue/red LEDs along the front that indicate charging, Wifi/Bluetooth Status, etc... look good as well.

The back looks cool imho, but I really do wish there was some way to turn off the LED.

Physically I think it is nearly identical to its predecessor, the G51J. Internally it has a number of differences, including the GS360M GPU with its GDDR5 memory and instead of a second hard drive bay, the G51Jx features two additional RAM slots to bring the reported maximum RAM up to 16GB.  

I have to admit though, the size of the battery surprised me - the 6 cell it came with does not stick out at all but sits flush with the system - but it is so small and light compared to other laptop batteries I've used in the past and especially considering how big this laptop is that I think they could have designed things to fit in a larger one could definitely have helped out with more battery life. Otherwise though the system seemed as it should.

Note that the side mounted Express Card slot holds a dummy ExpressCard with space for a spare MS and SD cards to be stored if you want - however there is no interface to actually use the cards in that position. The front 8-in-1 card reader though works well for this. I plan on eventually utilizing the Express Card slot to add a USB3.0 port or two to my G51Jx when I get some peripherals that support it - they are available now, but they are fairly bulky imho. Hopefully in future they will be able to offer models that are flush (or nearly flush) with the rest of the system.

As with the G51J the Power Brick is a pretty substantial brick - it's 120W and rather bulky. It hot fairly warm during my tests, but never actually hot.

Back Panel

G51J/G51JX lid

I do think the back panel looks pretty good in real life. It's quite a fingerprint magnet, but the design is impressive. The panel is identical to the G51J's one from all the pictures that I've seen. One thing I don't like too much is that the white lights on either side of the panel cannot be turned off without going to battery mode - and the central light (under the RoG logo) can never be turned off. This could be potentially annoying if you leave the computer on overnight downloading, recording, encoding or something while you're trying to sleep nearby. It does look pretty cool at other times though - I just wish we had some control over when it's like that.


I like that there are multiple options for the illumination of the keyboard - I had thought there would be only on and off. I find that the keyboard has some flex and clatters a fair bit.


I like the Touchpad surface, but not the buttons. They are difficult to press, especially the right one. They work, but it isn't comfortable imho. Definitely not going to disable tapping. I definitely like that there is an easy shortcut to disable the Touchpad (FN+F9).


The speakers on this laptop are actually a fair bit better than the Z96JS ones. At max volume the Z96JS speakers sounded somewhat tinny - while these were still true to the audio. The volume was also slightly higher - but not by much - almost all laptops have a lower maximum volume than most people would like. These speakers are ok to show videos to friends while on the go, but if you want real bass you'll need a good speaker system outside of them (this laptop sounds great hooked up to my Logitech X540 5.1 Speakers).

My PX200s headphones work great though and the volume levels when hooked up through the headphone port are excellent - certainly loud enough for anyone. And if you need either the speakers or headphones to be louder you can enable a pre-amp of course, but the distortion this causes depends on how loud the audio file was to begin with so it certainly varies.

All in all I feel the audio on this laptop is above average when rated against all laptops that do not have a subwoofer.

The Internals

My G51JX's Bottom

12 Phillips Screws (4 large and 8 small) to open up the case to clean the insides or change out the Hard Drive or RAM - but at least you can look everything over with that one panel. The Z96JS I had before had a similar setup, but without quite as many screws. One of these screws was in quite deep and needed an extra long thin necked screwdriver to get it out (or in my case with my Multi-tip driver I pushed the magnetic screwdriver tip as far down the driver as I could and this was just enough). Note that the screwholes labeled K and the one with a screw image have different screws - they looked the same to me, but wouldn't seal, when I switched them it went in right away.

G51J's Bottom

Interestingly enough, you can compare the G51J bottom (above) from the previous revisions of the laptop - you can see there is no longer that huge round vent over the fan, yet the Fan is still in the exact same position, but there is no vents right over the fan itself - the only things I can see this being good for is possibly to reduce dust and to reduce noise... - maybe. It definitely is different though.

G50V Orange Vent

There are reports that the vent in the G51 and G51J may have degraded to just being decorative, being a molded plastic design instead of an air vent. For reference, that vent had been on all the G series laptops since the G50V when it actually was a vent - and it also actually orange...

As you can see from the picture (within the orange box), the internal structure of the laptop was designed for a second HDD - there is lots of empty space in that compartment. If they wanted to, based on the dimensions of the RAM slots and RAM they could have fit four extra RAM slots in the HDD compartment. Of note, the brand of the RAM cards are Kingston 2GB 1333Mhz ones.

You can also see the third antenna connection that is not currently attached to the default Mini Wireless Card (red box).

Assuming that the Heatsink works the same as the Z96JS one did then it goes from the CPU to the GPU. It looks to me like the GPU is inserted into a side loading slot and could potentially be replaced, like an MMX type card (blue box). I think I'm going to do some googling now to check this out - can anyone corroborate this? Even if it isn't a standard type, it's interesting that the GPU can be removed and potentially swapped if a unit with similar thermal/power requirements can be found and the BIOS could be made compatible with them.

It's interesting taking that into account, since it suggests that one of the only real changes between the G51J, the graphics card, may have been an easy change - and the RAM being in the position it is may be because that is how the G60Jx motherboad was set up - it would be interesting to see if the only difference between this and the G60Jx is the slightly larger (but much lower resolution) screen.

OS and First Boot

To get to the instructions screen on first boot was quick - however after letting the computer set itself up, it took about 8 minutes before it got to the 'Personalizing Settings' screen, and then a few seconds later it played the ASUS intro movie.

I won't even be keeping this Windows 7 Installation since I need some of the Pro Features for my web work - but at least it gives me a chance to test all the drivers and bloatware (which, fortunately enough, there isn't too much of) without having to install it all from scratch.

As noted in G51J reviews, the font DPI settings for Windows are set to 133% by default - just change that to 100% and you will be able to use the 1920x1080 screen as it was meant to be - and if you think that's too small you can change it back or bump up the DPI even more.

I like how it should be possible to run XP on this system as a dual boot option with all the drivers just in case. That was one way to get around the BSOD issue until it was fixed.


It has been noted elsewhere that the card is an Atheros one and the G51Jx has 3 antennas. The laptop has a button on the front reminiscent of a Thinkpad that lets you turn off all Wireless usage with a flick of it the switch. You can also turn on and off the Wireless devices through a hit on Fn+F2. Unlike with previous versions of the ASUS software, you need to click the individual wireless services if you want to turn just one or the other on - not just go through the options with the Fn+F2 key.


The retailer I bought from just says: RAM: 6GB DDR3 4 SODIMM Sockets

The outside of the ASUS brown box says: DDR3 1066 2G*3

But the RAM cards themselves say: Kingston ASU1333D3S9DR8/2G.

Here's the CPU-ID for them:

I have not changed any of the components in my machine - I started testing straight out of the box when I got it last night. At the retailer the box was sealed and I only opened it when I got to my car. There's an investigation to see if only I got lucky, if there was a mix up or a supplier changed the stock. It's important since other retailers are charging almost $200 over the price of the original RAM to change the modules from 1066 Mhz to what I got with my system at no extra charge.

For example:
6,144MB (6GB) DDR3 1066MHz Dual Channel Memory (2GBx3) [Included]
6,144MB (6GB) DDR3 1333MHz Dual Channel Memory (2GBx3) (+ 189 USD)

Depending on the outcome there could be some very happy people out there .

Update: All users who checked have found that their G51JX units have DDR3 1333Mhz RAM instead of the advertised DDR3 1066Mhz modules. However, it has been reported that ASUS has done this because of a Supply Issue and the 1066 Mhz modules were not available in sufficient quantities. There is expected to be a mix of G51JXs with the enhanced memory and those without it. The box will list the official 1066 Mhz stats, there is no way to know which you have without opening it up and checking physically or with CPUz or other similar applications.


Interestingly, if you disable Quiet Boot you will notice that when the computer boots, it says "American Megatrends Evaluation Version." The BIOS version I have is 204.


Putting my hand to the vent on the left side of the system it has never felt burning hot - but very hot definitely. It cooled down fairly quickly after the stress is reduced.

Idle Temperatures of the system, about 20 minutes after boot withe minimal actions being performed by the system and 22*C ambient temperature and all stock cooling:


The fan was not audible at first, but when under load (3DMark06) it was - but it wasn't too loud. Will get more details on this and the heat as time goes on.

Performance Tests

Windows 7 System Assement

I ran this with Extreme Turbo enabled - the only thing that changed is the Processor score went from 6.9 to 7.0 - otherwise this is all stock. The free DDR3 1333MHz upgrade really helps take the RAM to the top score of the system (along with the fact there are 6GB total, of course) .

Here are my 3DMark06 scores with Extreme Turbo enabled:

And 3DMark06 without Turbo:

Both these tests were run at the default clocks from ASUS with no overclocking at all, the resolution was 1280x1024 and all settings were 3DMark06 defaults.

The nVida Drivers my system came with (and which I had on when I did the test) are 187.66. I did these tests on the first boot and I didn't install anything before doing the test, though I had about 5 IE windows open at the time.

As you can see, there is a 205 point change with Extreme Turbo On specifically for the CPU score - but there is no effect on the other numbers, which stay well within a reasonable margin of error/test fluctuation. So Extreme Turbo really does appear to help on the CPU side of things - with a bump up of just over 15% in this test - and if we overclock the GPU some we should be able to bump things up there as well . I repeated this test about 5 times and each time it follows that the CPU numbers without it are ~200 points lower.

3DMark Vantage w/ Extreme Turbo Enabled:

I ran that right after rebooting the system - still with all the same bloatware and drivers as the laptop came out of the box with.

3DMark Vantage w/ Turbo Disabled:

I ran this right after the Extreme Turbo one. It appears that the CPU score was unaffected, and actually went down a bit in this test, which contradicts the results from 3DMark06 but matches some G51J CPU tests I recall reading.

For example, here is a sample of the 3DMark06 tests to make sure the CPU Turbo vs. non-Turbo was not an anomaly:

9652 turbo: CPU: 3176
9527 no turbo: CPU: 2985
9634 turbo: CPU: 3140
9621 turbo: CPU: 3129
9516 no turbo: CPU: 2967

Update: Thanks to thalanix's suggestion I manually edited the setup files (nvam.inf and Listfiles) to allow me to install the latest nVidia drivers, namely: 195.62. Here are the 3DMark06 results for that:

No significant change unfortunately - but next step - overclocking will hopefully be more promising .


The GS360M GPU should only be using 50W by default as opposed to the 65W of the GX260M in the G51J - so hopefully there should be no throttling issues of the PSU not giving enough Watts. Also, since the GS360M is 40nm it should run cooler at the same performance levels and hopefully overclocking potential will be even greater - but I know what you're saying, "Stop talking and start testing!" - I will, I will - check back tomorrow .

I tried to upgrade to the latest nVidia drivers, but currently the GeForce 300M Series (Notebooks) is not an option on their site - just 100M and 200M. Doing a manual scan through the nVidia site came up with the statement that "the manufacturer of this system requires that you download the driver for your GPU from their support site." Will try download and installing the 195.62 Notebook drivers through the 200M link - here's hoping it doesn't do a check on the Device ID and decide that the 300M isn't yet supported in their latest final drivers (which are from December 4th 2009) or latest beta (December 15th 2009) - feels like the release of Vista all over again... Here's hoping they get back from Christmas vacation soon.


Confirmed, the latest nVidia drivers as of today (Feb 4/10) will not install. On starting the installation the installer detects that "no compatible hardware was detected" and setup exists.

This could be a promising sign though - maybe there are some enhancements to be added to the drivers that could make the GS360M do even better - or they'll just add in the Device ID with no customizations - only time (or perhaps a leak) will tell...

I checked the ASUS website and it does have a series of files for the G51Jx, but unfortunately they are all versions I already have, including the 187.66 Graphics drivers (which appear to be from August) and the BIOS 204 that they added to the site less than 7 days ago. Hopefully updated drivers will be forthcoming soon. Until then I modded the files (as mentioned above) to do the tests.

Overclocking the GPU

Stock Clocks: Graphics Clock: 550 Mhz Memory Clock: 1800 Mhz Processor Clock: 1323 Mhz

Here is a temperature chart of a single run of 3DMark06 with all stock clocks and cooling as a baseline, ambient temperature was 22*C:

Run #1: Graphics Clock: 588 Mhz Memory Clock: 1845 Mhz Processor Clock: 1414 Mhz

Ok, I still need to stress test these values some more, but I thought I would do a pretty minor overclock from the stock settings to start with and then gradually work my way up.

I did not notice that the laptop got much hotter than usual when running this test. I was able to perceive though that the graphics were noticeably smoother in this test though.

So Overclock #1 broke the 10,000 barrier. I will up it a bit higher and then leave leave it overnight doing a loop as a stress test. Note that this is without Turbo being enabled.

Run #2: Graphics Clock: 600 Mhz Memory Clock: 1960 Mhz Processor Clock: 1443 Mhz

Extreme Turbo CPU Overclock Mode enabled

This time I kept up the proportions when overclocking, and I also got a temp reading right before and right after the 3DMark06 run. I watched the run closely and there were no artifacts I could see.

Before Temp:

After Temp:


Run #3 & #4:

I reached a wall - I was getting "IDirect3DVertexBuffer9::Lock failed: n/a (Unknown)" and "IDirect3DDevice9 Present failed: Device lost" errors - no lockups or BSODs or spontaneous reboots, but these errors would come up if I either overclocked the GPU too high or had the GPU overclocked above 615 with Extreme Turbo enabled.

So that's about as high as I can go with the OC without changing something. If anyone has any recommendations on other enhancements to try or suggestions whatsoever on this front, please let me know .

3DMark Vantage Overclocked:

The max temps of the CPU Cores for this run was 73, 67, 69 and 71 - GPU reached a peak of 85*C

Another test with similar clock settings that passed as well:

This has the highest CPU score I saw 1160+ above the previous one, and the GPU score differed by -26 points from the previous.

Other tests I ran as well:

1615/1990/1525 Passed - Score: P6096
1620/1990/1550 Failed (Error Message) - Max GPU 85*C
1615/1990/1535 Failed (Error Message) - Max GPU 85*C
1615/1990/1530 Failed (Lockup - Cursor in form of hand visible in middle of screen, but it will not move, fan is on loudly, no buttons respond - after 3 minutes with no change I shut down.)
1615/1990/1525 Passed - Score: P6122 - Max GPU 85*C
1615/1990/1525 Failed (Error Message) - Max GPU 85*C (These exact same clock levels had passed 5 times previously)
1610/1985/1520 (3x Looped for Stress) - Failed during first loop (Error Message) - Max GPU 85*C
1610/1985/1520 (3x Looped for Stress) - Also failed during first loop - (Error Message) Max GPU 85*C

The Error Message I received was:

It's not a very popular error message from what I can see. Could there be another explanation for it? It seems like that and the 3DMark06 errors may be coming up randomly and may not be signs that the overclocking is too much.

I think I will next try to revert to the standard ASUS nVidia 187 series GTS360M drivers and see if that has any effect.


I tried to install the latest DOX drivers to continue testing (http://www.laptopvideo2go.com/drivers). The driver would not install unfortunately, even though the default inf file did have the GTS360M listed.

So with the standard ASUS drivers I was able to get a higher score in 3DMark Vantage. I ran 5 tests and this is the highest that succeeded without any errors:

At 635/1990/1615 I received an error near the start of the first 3DMark Vantage test that the nVidia driver kernel had crashed. However, 630/1990/1600 was stable in 3DVantage for all the tests I tried - but when I went back to 3DMark06 I was getting errors with the clocks set that high. Reverting to 615/1990/1550 solved those problems, and I continue to test things out...

Safe and Stable GPU Overclocks w/Extreme Turbo Enabled:

From all my testing here is what I consider to be both safe and stable overclocks for my particular G51Jx w/3DMark06 and 3DMark Vantage w/Extreme Turbo Enabled: 610Mhz, 1980Mhz, 1530Mhz. These clocks result in approximately scores of 10,200 and P6200 respectively. Even under the stress of multiple runs of these benchmark apps, the GPU clock has never gone above 85*C w/22*C ambient temperature and none of the 4 CPU cores passed 5*C. No artifacts were ever observed even at a higher overclock - but this one is what I would consider safe and stable personally. I ran the tests 9 times consecutively on both pieces of the software with a problem. I'm sure more is possible depending on the game and the particular system - but having this as an option that stays that cool should be great and help keep the laptop lifespan from getting reduced too much.

Boot Times

These are all after the first boot and are measured from the moment I pressed the power key to when I was at the desktop (not including the time it took to select my account and enter my login password)

Default Windows x64 Home Premium on AC: 42 seconds

Default Windows x64 Home Premium on Battery: 1 minute 5 seconds

Battery Life

None of these tests are synthetic but are real world usage experiences.

Battery Life Test #1: Me working on this review, checking e-mails installing Firefox and other mundane computing chores while on Battery Saving w/max 33% processor, 49% screen brightness, Wifi on. At 25 minutes into the test there was a reported 76% battery left and 1 hour and 35 minutes remaining. At 1 hour 45 minutes from when I hit the power button, the system went to Hibernation with 3% battery remaining. The system was pretty much silent and the big vent on the left side was only just warm to the touch.

The Battery Life of this series of machines is known to be quite low, since they are built for speed more than anything else, but at least they still provide the service a built in portable UPS and if you can get 2 hours of light work and installations with the default 6 cell battery then I'd rank that as acceptable to me. If I want more I can bring on of my AC PowerPack/PowerPlugs.

Power Brick/Travel Power Adapter

While testing out the G51Jx I looked at the power plug and I thought it looked like the same size as the one on my ASUS Z96JS - and so I brought out my Kensington Universal 120W AC/DC Power Adapter that I purchased to use my Z96JS in cars and airplane with the DC plug and also at home through the AC port.

I'm currently running a test to make sure there is no throttling and that the portable 120W adapter gives enough power for the G51Jx under load. With my Z96JS I was sure I was safe since it was a 90W laptop - but since this is 120W only, I'm doing a full test cycle just incase.

For more info on the Travel Power Adapter, Tip ID, etc... click here to go to the Compatibility Report.


I wish there had been some option with my retailer to get a Blu-Ray Drive added at point of sale (instead of searching for aftermarket ones), but otherwise I'm otherwise happy with my G51Jx. I managed to get a price match with another retailer that supplies other retailers and had to order from their Warehouse and so I got the unit in my hand within 2 hours of deciding to get it - if it wasn't that I spent a week considering the G51J/Jx before hand then this would have been insanely compulsive for me - but I think I considered everything and if this thing works as expected it should be awesome . They were selling out fast in all the various places I checked in my local area (Toronto), especially caused by sub-retailers buying up the chain and raising the price along the way - so if you want one, I suggest you call up a store and put one on hold now. From what I heard the next wave of shipments is at this point scheduled for Feb 28th or so.

Probably the only reason I got mine without putting in a reserve or pre-order is that people think the G51Jx won't be out for another couple of days - but it's out there right now - so if you want it I'd do a search right away and you might come up lucky like me .

This is definitely a work in progress - there's still a lot more to test, like Battery Life and more on Performance, but so far, so good. I plan to update this post as time goes on - if you have any questions on these results in particular or want me to test or take a picture of something else to add please reply below and I'll see what I can do . Either way though, I hope you enjoyed reading my G51Jx Review - I (and some others) have written a lot more for hundreds of other devices as well, you can check them out on the rest of the rest of this site if you like .


Last Edited By - PJPeter on 2/14/2010 11:02:46 PM

 Posted - 2/4/2010 11:14:49 AM  - Show Profile  Edit Message

3 Contributions

Hello, That is a nice review - I was thinking of it beforeand this helps but I am still not sure. Are the speakers as quiet as other G laptops? Please post pic of inside too.


 Posted - 2/4/2010 11:28:09 AM  - Show Profile  Edit Message

3 Contributions

Nevermind I reread it and you answered this already...

Last Edited By - Talismanna on 2/4/2010 11:29:33 AM

 Posted - 2/4/2010 11:31:23 AM  - Show Profile  Edit Message

Site Administrator
2020 Contributions

Sure Talismanna,

Happy to hear from you . I will open up the system and take a couple of pics as soon as I finish with my battery life test.

As for the speakers, I usually use PX200 headphones and when I want to share with external speakers I hook up my Logitech X540 5.1 Surround Sounds - but it's true if I'm at a friend's place or away from my desk something I definitely would want to use the onboards, and I'd want them to be loud enough or there wouldn't be a point. So I'll transfer over some mp3s and give it a test in a few minutes and I'll add a "Speaker" section to the review - thanks for the suggestion .


 Posted - 2/6/2010 10:38:50 PM  - Show Profile  Edit Message

Site Administrator
2020 Contributions

I was sent this private message and I requested that I be able to answer it publically so everyone can see the discussion and possibly join in.

"Hi. Congrats on your new laptop. Love your review blog!

Could you tell me in more detail:

1) how many cooling fans does your laptop have?
2) how loud is the fan while gaming (or while the CPU and GPU are under a heavy load? Is it loud enough for people in the same room to notice?
3) does it seem like the fan is always on or on most of the time under stressed load?
4) does it seem like the fan is always on or on most of the time under regular load?
5) how hot does the laptop get on the BOTTOM when the CPU and GPU are stressed out? Is it searing, painful to touch hot? Or just really warm?
6) Do you have concerns about frying the CPU or GPU while both are under a heavy load?

Thanks for your help!

Here is my response:

1. 1 I believe
2. It gets loud, but not too loud. I can give you a db reading at 3 feet or so if you like.
3. and 4. Yes, when stressed the fan goes to level 2 and it is noticeable. Then under stress (a few seconds into 3DMark tests for example) it goes to level 3 and stays there until a few seconds after the test ends and then returns to level 2 for about 30 seconds and then back to level 1. Level 1 is the cooling under regular loads and it is not quite silent but also certainly not noticable unless you concentrate on it. That level is a little quieter than my Z96JS is on it's lowest fan level. I don't know if the fan goes off - I'll have to let the system idle and see, but it seems it stays on at that near silent low level all the time.
5. The bottom does not get searingly hot - it does get quite hot, but only on the left hand side of the laptop where the big vent is. Otherwise the rest of the laptop is cool to the touch. It gets hot on both the top and bottom on that side, but only above and below where the GPU and CPU (and their associated heatsink) are located - it gets a bit hotter on the bottom - but not by much. It is not painful, and this being winter where I am I actually am enjoying touching it :p. In summer it may be hotter and not as pleasant since the ambients will be higher, but right now it's quite nice :).
6. No I do not have any such concerns. So far the system has been stable and the GPU has never gone above 85*C in rigorous testing (that would usually send a G51J's GTX280M up to 90*C or more) and, though I haven't tested it as vigourously, I have not seen the the CPU temp peak at anything higher than 74*C. I do have a 3 year warranty though so that also makes me feel better as well (2 years ASUS and 1 year Visa) :).

To expand on the temperature issue, you can get a cooler - I prefer not to have one since I don't like increased noise. I actually have a passive cooler on it right now, it's simply a Targus lap desk with a wavy design on top and many vent holes to allow air to escape. It means when I put the laptop on my lap it is not directly against my skin or clothes and so has better ventilation. On the desk it probably isn't helping that much, but it can't hurt either.


Last Edited By - PJPeter on 2/6/2010 10:40:17 PM

 Posted - 2/14/2010 5:07:23 PM  - Show Profile  Edit Message

Site Administrator
2020 Contributions

A video review was posted a couple of days ago:

I assumed there were so many G51J physical reviews, and since the outside of the G51Jx was identical I didn't bother with going over the same ports, design etc...

But if you want to see the temps, weight and so on in a G51Jx in particular, this vid is for you :

Note that, even though the video says the eSata port doubles as a USB one it does not. That is a special feature on some Hp and Dell notebooks - unfortunately it does not exist on this ASUS one. If it did it would be advertised. Plus, I just spent 5 minutes trying to get either a standard USB-A or a non-Standard USB plug to fit and none of them worked. Wish it was, but it isn't a dual port unfortunately.

He also says you can control the lights, something you cannot do at this moment, and he says the center horizontal chrome plastic bar on the back of the lid lights up - I have yet to see this happen after two and a half weeks of ownership.


 Posted - 2/14/2010 10:45:35 PM  - Show Profile  Edit Message

Site Administrator
2020 Contributions

It's only now, a week and a half from when I finally got a few minutes to try out the free accessories. The free backpack is a Green/Black model with the ASUS and ROG logos on it.

I've found it pretty good - but it was not designed for a 15.6" notebook. It was maybe meant for a 15.4" - I've found with the rest of the pack fully loaded, I really have to work to make the backpack close. Also, the padded section doesn't cover about 1" of the laptop - not an optimal carry case for the 15.6" G51Jx-A1.

I had been expecting to get a blue under-lit ASUS branded Razer Copperhead Mouse. What I got was a Razer Salmosa. It has no light and only 3 buttons - though it does have adjustment toggles on the bottom to change the dpi and hz settings, is a lightweight laser mouse and the bottom is Teflon smooth which slides over the mousepad as if there is almost no friction at all.

I've found though that it is too low and so it doesn't allow me to rest my fingers over the buttons comfortably. Also, going from the Logitech MX518 Gaming Grade Mouse with its 7 fully software customizable buttons with adjustable dpi switches on top - I can't really take it. I'll be putting it back in storage - if anyone wants it, feel free to post.

I originally thought I would need to install a Professional version of Windows 7 - however, I found that you can in fact install IIS on Windows 7 Home Premium x64 (something many other articles deny). So I'll be sticking with my default W7 install - making a backup disc right now .


Last Edited By - PJPeter on 2/14/2010 11:08:08 PM

 Posted - 2/17/2010 1:11:21 AM  - Show Profile  Edit Message

Site Administrator
2020 Contributions

There are reports that other users are receiving Razer Abyssus Mice instead of Salmosas - though the Abyssus also only has 3 buttons, it has a higher resolution sensor and is approximately .5cm taller. For me, going from a 4.2cm Logitech MX518, the 3.5cm Salmosa is just too low for me. No matter how long I use it, it is just not natural for my hand, especially with my Belkin Wrist-rest Mousepad (which is what's standing between me and a recurrence of Carpal Tunnel). Certainly the Copperhead and Diamondback are superior mice in every way that counts to me imho - but the Abyssus is at least better than the Salmosa in the comfort department if nothing else...

There is a CPU Throttling effect if you are utilizing both the CPU and GPU at 100% as with a FurMark Extreme Burn test for example. If you have the screen brightness on max and the GPU overclocked or at stock clocks then you will notice the first CPU core does not go into Intel Turbo Mode at 2.8GHz, instead staying at 933mhz because it is starved for power. With FurMark running this test I was able to get to 86*C on my GPU - but no higher. If you underclock the card and turn the brightness down low, then you can see Intel Turbo Mode being entered.

Here are some tests comparing the G51J Overclocked to the max before throttling took effect vs. the G51Jx Overclocked until 3DMark Vantage would start to give errors. Note that the G51J was using the latest drivers and had the x0124 BIOS fix for BSODs, while my G51JX is completely stock, including thermal compound and lacking a cooling pad and using the default 204 BIOS and with the default nVidia 187.66 drivers (the only ones available for the GTS360M at the moment, until the official drive release from nVidia for this model in early March).

The G51J results were posted by thalanix. On 3DMark Vantage with 580/1450/975 clocks with No Turbo, No Physx. as high as it could be pushed without significant throttling. Card max is ~600 on the core. The temperature of the GPU maxed out at 89*C during the run. The final score was G51J GPU 580/1450/975 clock with No Turbo, No Physx P5777 3DMarks, CPU Score: 10647, GPU Score: 5012.

Also there was another previous G51J test with the same system: G51J GPU 550/1375/950 clock with No Turbo/No Physx: P5636 3DMarks, CPU Score: 10597, GPU Score: 4875

The results for my various runs are below:

No GPU Overclock (eg. Stock ASUS Clocks Graphics Clock: 550 Mhz Memory Clock: 1800 Mhz Processor Clock: 1323 Mhz) with PhysX and Extreme Turbo enabled

With GPU overclocked and Turbo and PhysX disabled

With GPU overclocked and Turbo disabled and PhysX enabled - did an identical run with PhysX disabled and got: P5854 GPU: 5128 CPU: 10172.

With GPU overclocked and Extreme Turbo and PhysX enabled. I did this run to show that Extreme Turbo was not causing significant throttling. Also did an identical run with PhysX disabled and got: P5855 GPU: 5110 CPU: 10404. Note there is a difference between the GPU and CPU levels in each run - but I don't know that the flux is more than within the standard difference between runs.


Last Edited By - PJPeter on 2/17/2010 2:03:34 AM

 Posted - 2/18/2010 2:58:03 PM  - Show Profile  Edit Message

Site Administrator
2020 Contributions

GPU Overclocked (630/2000/1600) + Extreme Turbo + Full Brightness = 3DMarkVantage Crash or Error Stop after Jane Nash and most of New Calico and before CPU Test 2 - Not sure what moment it happened as I was distracted, but it happened twice at approximately the same time. I rebooted after each crash, since the performance was lousy after a GPU Driver crash and reset.

Previously I was testing with lower than max brightness and it was fully stable over 10 trials at this clock and slightly higher. I had it on a lower brightness because that was comfortable for me, 100% is too bright imho - but I will now test using Max Brightness when I do stability testing because of this throttling.

Last week I noticed that the G51Jx would crash and reset the GPU 3DMark06 on anything over 615/2000/1600 and later it could go higher. That might have been why.

I have now re-run the test at 0% brightness P6433 - GPU:5123, CPU:27602. Running it again and will keep upping the brightness until I see a crash again.

Second Run: 0% brightness: P6472 - GPU:5148, CPU:28271
Third Run: 7% brightness: 3DMark Vantage Crash at end of CPU Test 1
Fourth Run: 0% brightness: P6430 - GPU: 5108, CPU 28763
Fifth Run: 7% brightness: Same as before: 3DMark Vantage Crash at end of CPU Test 1

This is Performance being limited because of Power available to the GPU/CPU imho - one or the other chokes and causes the GPU Driver to crash. I'm curious now which actually causes the crash, the first CPU test or the very end of New Calico - will test again, and again, and a few more times after that to confirm exactly when it's happening and what else I can see about it.

Some more results:

Sixth Run: 93% brightness: (Fresh boot to ensure previous low brightness run is not biasing the data) - 3DMark Vantage Error: "Check Failed!...SetForegrondWindow failed" Error at end of CPU Test 1.
Seventh Run: 7% brightness w/No GPU Overclock (Also Sanity Check): P5499 GPU:4350, CPU:26516
Eight Run: 0% brightness w/GPU Overclock: P6403 GPU: 5099, CPU: 27487
Ninth Run: 26% brightness w/GPU Overclock: Crash before CPU Tests - 3DMarkVantage black box for 30 seconds, then 3DMarkVantage gave DXGI_ERROR_DEVICE_REMOVED when it was restored (Nvidia GPU Driver Crashed)
Did a part of a test run to see how well drivers would respond after crash, max fps in Jane Nash was 5 fps, average ~2 fps - down from 22 max, ~17 fps average
Tenth Run: 40% brightness w/GPU Overclock Just ran CPU Test 1 and 2 - No crash
Eleventh Run: 40% brightness w/GPU Overclock and all tests: 3DMark Vantage Error: "Check Failed!...SetForegrondWindow failed" Error at end of CPU Test 1.

Every test with my 630/2000/1600 overclocks pass as long as back light is set to minimum. If it is set any higher, then it gives either GPU driver crashed (three times) or 3DMark SetForegroundWindows failed error (all others). I thought I was wrong when it passed once with the GPU Overclocking and CPU Extreme Turbo both enabled and the brightness at 7% - but it turned out that P4G was being stupid and whenever I hit Save it was showing 7% but the actual display brightness was 0% (seen by trying to go down and getting no difference in brightness, just the "Brightness Down" display coming up and ticking down a bar, but with nothing else happening). When I set P4G high enough that it did actually save as a display brightness above 0%, the same error as on the other runs re-occured.

So if the display is set above 0% I get errors consistently with GPU Overclock and CPU Extreme Turbo enabled that I do not get if either of those are disabled and the backlight is set to max. If this is not evidence of some kind of power starvation, I don't know what is...

I'm satisfied through over a dozen full runs that this holds true - it'd be great if someone else can confirm it as well but on my system as it is today, I think it's a fact.


Last Edited By - PJPeter on 2/18/2010 3:01:38 PM

 Posted - 2/18/2010 3:00:45 PM  - Show Profile  Edit Message

Site Administrator
2020 Contributions

Test with GPU Overclocked (615/2000/1600) + Extreme Turbo + Full Brightness with rx1950 PDA (5v ~2000mA w/Charging+Wifi On+Full Screen brightness) + 1 USB Plasma Balls (5v ~500mA) + USB Flashlight (5v ~100mA) + USB Vacuum (5v ~1A on full speed).  I have confirmed the USB Power requirements of the Plasma Balls and PDA in power tests I recorded on Stuffopolis.net. Therefore the power draw was 15-18 Watts from the min to max requirements from my power tests on the devices.

The test was stable and the results over two runs were: P6223, GPU:5044, CPU:20782 / P6245 GPU:2063, CPU:20811

With no devices connected to USB results over two runs were: P6327, GPU:5045, CPU: 26592 / P6330, GPU: 5550, CPU: 26418

With 20% brightness with lots of devices connected to USB: P6289, GPU:5048, CPU:23972

With 20% brightness with no devices connected to USB: P6357, GPU:5063, CPU:27189

One other interesting thing I found was that the USB ports on each side are individually powered. If I have the Vac plugged into the same side as the rx1950, the rx1950 will show signs of power starvation (flashing screen from AC to Battery). If they are plugged into opposite sides of the laptop, then they can both work with full power.

Comparing the CPU score with no extra USB devices connected vs. with there is a consistent 3000 point difference with the screen brightness set for 20%.  

With Brightness set to 100% the CPU score difference is over 5500 points. That's a 21+% drop in CPU Performance.

For both sets of runs, the GPU Score was identical - it's only the CPU score that is being choked by the lack of power

This is with PhysX enabled.

There was no crashes as there were when I ran the tests previously with full brightness and a higher GPU overclock.

To be completely methodical and make sure that it was not Windows Mobile Device Center that was causing a difference in scores I re-ran the test with the rx1950 fully charged with it's Wifi Off and Screen brightness turned to 0. (Note that the Bluetooth and Wifi were both enabled and the Screen Brightness were all turned to max on the Notebook). With these settings the 3DMark Vantage results were: P6407, GPU:5112, CPU:26722. So that had no effect at all on the score differences. It is the USB Power Drain combined with the Brightness that is causing the CPU to under perform.

This is in addition to the tests showing consistent crashes at more than 0% brightness yesterday. These two tests together show in two different ways that throttling is taking place.


Last Edited By - PJPeter on 2/18/2010 9:22:39 PM

 Posted - 2/19/2010 1:40:24 AM  - Show Profile  Edit Message

Site Administrator
2020 Contributions

To test the hypothesis that the CPU was being throttled because the GPU was helping it through PhysX:

I went into these tests agreeing with this, since it makes sense that with the GPU trying to help the CPU with the calculations via PhysX it would be robbing power from the CPU and causing the throttling. Hence I thought the throttle should disappear if I re-ran the tests.

No PhysX - P4G Extreme Turbo Enabled

USB Devices and Setup:
1 USB Hub on each side of Laptop
Hub 1/Side 1: rx1950, Fan 1, Fan 2, USB Flashlight, Plasma Ball 1
Hub 2/Side 2: A696, Xperia X1, Fan 3, Plasma Ball 2, USB Mug

Brightness 100% 615/1990/1590 with No USB Devices: P5749 GPU: 4998, CPU: 10470

Brightness 100% ASUS Stock GPU Clocks with No USB Devices Attached: P5132 GPU: 4397, CPU: 10295

Brightness 100% 605/1980/1580 HP rx1950 PDA + Sony Xperia X1 + ASUS A696 PDA + 3 USB Fans + 2 USB Plasma Balls + USB Flashlight + USB Mug with built in heater: P5202, GPU:4966, CPU:6065

Brightness 100% ASUS Stock GPU Clocks HP rx1950 PDA + Sony Xperia X1 + ASUS A696 PDA + 3 USB Fans + 2 USB Plasma Balls + USB Flashlight + USB Mug with built in heater: P4744, GPU:4379, CPU:6324

Brightness 0% ASUS Stock GPU Clocks with HP rx1950 PDA (2Amp) removed: P5132 GPU 4391, CPU 10388

Brightness 100% ASUS Stock GPU Clocks with HP rx1950 PDA (2Amp) removed: P4845 GPU 4386, CPU 7060

Brightness 100% ASUS Stock GPU Clocks with Sony Xperia X1 removed: P4892 GPU 4418, CPU 7215

Brightness 100% ASUS Stock GPU Clocks with Sony Xperia X1+rx1950+A696+Flashlight+MugHeater removed: P5147 GPU: 4406, CPU: 10402

Brightness 100% ASUS Stock GPU Clocks with All USB Devices removed: P5118 GPU: 4376, CPU: 10419

I ran each of these tests through at least twice. I was sometimes receiving the 3DMark Error messages occasionally (that is why I changed the clocks slightly during the early tests, and then preferred Stock Clocks later on), so I even more heavily think it has to do with power throttling on the CPU. When the GPU gets throttled, 3DMark gives an error and there is a GPU driver crash.

It seems that even just at GPU Stock Clocks and without PhysX there is a large effect on the CPU score (~40%) with various USB devices attached at different brightness levels. This is in comparison with the 21% difference when PhysX was enabled - the total CPU score was more than twice as much, but the difference in CPU score caused by this throttling was only ~1/3 greater.

I used the fans and my PDA and Phone instead of the USB Vacuum because that Vacuum is too darn loud - it can use up 2A @ 5V easily. The rx1950 requires 2Amps at the max settings I had it set for, the A696 around 500mA, each of the Plasma Balls also 500 mA, the Xperia maxes out at about 800mA, the Fans are all quite variable, but use about 400 mA average over the three and the USB flashlight uses a steady 100mA to recharge its internal NiMH battery (it's a dumb and hence slow low amperage charger). So the total load was around 5.6A at 5V if all devices were requiring their maximums. Since I'm not 100% sure on some of the mA requirements (especially for the fans) I will give it a large range - so the total Watts for this setup was approximately 20-28W which was higher than the estimated 15-18W in the first test.  You can see from the tests that taking away one or the other loads would affect the CPU score between 10-12%, so with that much load it was near the tipping point, even though the GPU was only running at stock clocks. You can also see the tip when lowering the brightness to 0% and taking away 1 part of the load was enough to bring the scores back up to full normals.

So taking the estimates from thalanix's post: 45W CPU + ~22W GPU at Stock Clocks + ~20W USB load = 87W without even taking into account the screen, internal fan, HDD or any other power requirements of the laptop. It would make sense at that sort of load that there would certainly be throttling. In this case the GPU is not helping with PhysX, but it is certainly not entirely idle during the 3DMark Vantage CPU Tests.

The load was much greater than the previous PhysX test so it doesn't mean that the hypothesis was wrong - it could have been that PhysX was causing the problem to be visible at a lower USB load amount. However, it is visible without PhysX and any GPU overclocking, just using power the USB ports supply freely.

Note that I did not have a USB mouse attached during the tests, that could easily have substituted for the flashlight instead.

In the end, I ran out of USB ports and high powered USB devices - the G51Jx can power more things through USB - but as you can see, having even as much as I did will affect your performance.


Last Edited By - PJPeter on 2/19/2010 3:26:53 AM

 Posted - 2/20/2010 2:31:18 AM  - Show Profile  Edit Message

Site Administrator
2020 Contributions

I did a test of the G51JX's power requirements using my Walltap Power Meter through one full run of 3DMark Vantage.

G51Jx Idle 46W - Extreme Turbo+GPU Overclocks (though idle underclocks on both since system idle)

G51Jx Extreme Turbo + 630/2000/1600 GPU Clocks + Full Brightness (No USB Devices) w/PhysX disabled

I copied down the following numbers from the Power Meter while it was set to 'Watt Mode.' I tried to record all the numbers as they changed:

G51Jx Loading Jane Nash: 90, 85, 90, 82, 95, 93, 92, 90, 95, 94, 91, 89, 90
G51JX 3DMark Jane Nash: 97, 108, 99, 102, 104, 105, 109, 96, 107, 101, 111, 114, 115, 108, 107, 98, 103, 114, 100, 110, 101, 98, 103, 98, 94, 95, 98 End of Test: 69, 75, 86, 90
G51Jx 3DMark New Calico: 94, 101, 95, 94, 100, 94, 98, 101, 99, 102, 93, 90, 104, 92, 96, 104, 101, 96, 92, 94 End: 76, 78
G51Jx 3DMark CPU Test 1: 104, 103, 102, 103, 102, 103, 104 End: 69, 58, 85
G51Jx 3DMark CPU Test 2: Warm Up: 99, 102, 100, 103, 99, 95, 76, 105, 101, 103, 102, 85, 80, 102, 101, 103, 99, 102, 86 Test: 80, 82, 83, 82, 78, 80
Result: P5838 GPU: 5087 CPU: 10482

After test: 73 gradually down to 46 then 44 after the fan went to lowest spin, 42W-41W with screen dim
Monitor off, system idle: 37W

Note that this Power Meter is between the AC Adapter and the wall - so it does not take into account how efficient the AC Adapter and how much power it is actually delivering to the notebook. Just the draws from the wall socket directly.

Too tired to look into the numbers, just reporting them as I see them.

Note that, to test the accuracy of the meter I hooked it up to a lamp with a 40W incandescent bulb and got a reading of 39W constant.


Last Edited By - PJPeter on 2/20/2010 2:32:29 AM

 Posted - 2/20/2010 11:58:38 PM  - Show Profile  Edit Message

Site Administrator
2020 Contributions

By the way, I'm not sure if anyone noticed this but by default ASUS underclocked the GTS360M in the G51Jx as they did the G51J's GTX260M:

The default clocks for ASUS are: 500/1250/800
The default clocks from nVidia are: 550/1375/950

The default clocks for ASUS are: 550/1800/1323
The default clocks from nVidia are: ???/<2000/1436

I've been able to overclock to 630/2000/1600 - I'm re-running my stability tests now with the power meter hooked up, full brightness w/my MX518 Gaming Mouse attached on both 3DMark06 and Vantage to see if this is in fact stable.


Last Edited By - PJPeter on 2/21/2010 12:41:34 AM

 Posted - 2/21/2010 12:41:46 AM  - Show Profile  Edit Message

Site Administrator
2020 Contributions

G51Jx Overclocking Stability Test w/Power Meters

To make sure the errors were not coming up because I had anything else open, I closed every window except the benchmark app.

3DMark Vantage was stable at 630/2000/1600 w/Full Bright + Extreme Turbo over 7 runs. However 3DMark06 was giving me Device Lost errors at this same clock speed. So I downclocked to 625/1575/1550 - still getting the errors. Below is a run at that level with me recording from the Power Meter in Watt Mode.

3DMark06 Run 625/1975/1550 + Full Brightness + Extreme Turbo + MX518 Gaming Mouse:

3DMARK06 Test 1 (Proxycon): 121, 117, 120, 125, 123
3DMARK06 Test 2 (Firefly): 121, 122, 123, Light up Log: 126
3DMark06 CPU Test 1: 90, 81, 98, 72, 78, 69
3DMark06 CPU Test 2: 93, 96, 99, 96, 99, 96, 71, 66, 77, 74
3DMark06 Canyon Flight: 121, 122, 123 >> IDirect3DDevice9:Present failed: Device lost (D3DERR_DEVICELOST) << 86, 67

Screen Off Total Idle After GPU Device Lost Msg: 31W, touching 29W min after 5 mins

System Off Wattage:

Shut Down Idle Wattage from PS alone: 3W constant

Stock W7 x64 Boot Wattages:

On boot: 4W, 53w, 73w, 64w, 73, 62, 50, 55, 57, 54, 55, 51, 66, 53, 40, 61, 54, 48, 44, 49, 48 / W7 Login / 47, 50, 88, 81, 77, 84, 66, 63, 80, 64, 73, 76, 62, 67, 86, 64 / W7 Idle @ Desktop / 61, 54, 42, 51, 38, 36, 33, 39 /Started 3DMark06, Ntune, P4G/ 69, 72, 90, 58, 60, 58, 40, 42, 38, 32

Change Brightness between 100% and 0% on Idle System: Flux from 35-38 -> 31->34

Changed the GPU Clocks down to try to avoid these problems:

3DMark06 Run 615/1950/1550 + Full Brightness + Extreme Turbo + MX518 Gaming Mouse:

3DMark06 Test 1 (Proxycon): 122, 117, 121, 119, 121, 122, 120, 119, 120, 122, 114, 117, 121, 118, 107
3DMARK06 Test 2 (Firefly): 120, 119, 122, 120, 124, 125, 122, 121, 125, 126, 91, 98, 95, 122, Light up Log: 126W
CPU Test 1: 94, 97, 98, 96, 92, 96, 97, 100, 70, 74, 79, 73, 67, 77, 76,
CPU Test 2: 85, 95, 98, 99, 95, 88, 70, 76, 70, 68, 73, 78, 74
Canyon Flight: 87, 100, 114, 120, 118, 121, 120, 121, 120, 117, 118, 119, 121, 112, 117, 120, 106, 108, 105, 109, 107, 109, 116, 117, 118, 109, 120 With Light up Jets and Smoke Trail on Air Ship Coming At You: 124, 122, 119, 118, 117, 112
Deep Freeze: 119, 120, 118, 120, 118, 119, 121, 120, 121, 91, 119, 118, 122, 121, 120, 122, Last Frame: 124
First 20 seconds @ Windows Desktop (w/Results): 85, 61, 43, 40, 43

Run Complete - 3DM06 Result Score: 10271 SM2.0: 4402 SM3.0: 4354 CPU: 3046

One other thing to add, I changed the Panel Scaling to "Do not Scale" and the Firefly Light up Log peak went down to 125W (which makes sense since less of the screen needs to be drawn). Previous to this it was using 'Nvidia Scaling.' With Light up Jets and Smoke Trail on Air Ship Coming At You likewise down to 122. Result: P10105 SM2.0: 4236 SM3.0: 4331 CPU: 3063 (within normal variance between tests).

I felt the Power Adapter after the 126W peak and it feels hot but the vent on the G51Jx was much hotter.

Will continue to run the G51Jx through 7 more 3DMark06 runs and then switch back to do 7 more Vantage runs before I confirm this as my stable overclock.

From what I can see, comparing with my previous 3DMark Vantage runs, 3DMark06 pushes the system more. The max Wattage in Vantage was 115 - here 126.

P.S. During one of the next 3DMark06 test runs I took out my battery to check the temperature - it feels slightly warm on top, but the bottom is cool to the touch.


Last Edited By - PJPeter on 2/24/2010 7:46:21 PM

 Posted - 2/23/2010 2:18:14 AM  - Show Profile  Edit Message

Site Administrator
2020 Contributions

Dell users have reported similar throttling issues to G51J/Jx users - ones that were even more severe up to 90W and then the same as ours after that. The solution to this was that Dell started sending users free 130W replacement Power Adapters and issued a new BIOS that allows them to be fully utilized. The performance of the systems increased dramatically. Turbo Boost was originally a big flop in all the reviews you read - a Single Core huge boost that was supposed to be a huge advantage that just never happened with most mobile systems. It seems this is because the manufactures have been starving us for power.

Some users suspect that performance may improve even more if 150W or greater PSUs are used since it would allow users to overclock their systems. Unfortunately the Dell BIOS has a hard 130W limit in it (as it previously had a hard 90W limit - and as our ASUS systems may have a 120W limit).

The affected Dell system is a Studio XPS 1645 with a 15.6" 1920x1050 FHD Display, Core i7, ATI Radeon HD 4670 - Not identical to the G51J/Jx - but a very similar Laptop.

I ran further testing to confirm if this issue is BIOS limited or PSU limited. Here are my results and info below.

Each of the PSUs are marked with Efficiency Ratings. The rating on my 90W ASUS Z96JS adapter is Class III. The rating on the 120W ASUS G51Jx adapter is Class V. Level III requires greater than 84% efficiency for a PSU w/Power Output between 51-250 watts. Level V requires greater than 87% for 49-350 watt PSUs. Level V also regulates the Power Factor remain at 0.9 or greater when under 100% load - which I reproduced in my test. The Kensington PSU with a lower level ratings went crazy with a PF of 0.66 or even lower sometimes, while the 120W adapter was almost constantly at 1 and never lower than .9.
(Reference: http://www.energystar.gov/ia/partners/prod_development/revisions/downloads/International_Efficiency_Marking_Protocol.pdf )

Therefore, assuming both PSUs go by their certification marking: The 90W adapter must take a maximum of 103.5W to deliver the marked 90W load. The 120W adapter must take a maximum of 135.6W to give it's 120W load (and 101W to deliver 90W).

I decided to be methodical and try many different settings and modes on both adapters - and so I took a couple of video recordings of the system and power meter w/each adapter to make sure that I had all the data and matched the Turbo Boost status and peak wattages at which things changed exactly.

The test was doing the FurMark Test (Stability, 4xAA, Xtreme Burn, Windowed 1440x900), using the Intel® Turbo Boost Technology Monitor + my Wall Power Meter with other values such as screen brightness and Extreme Turbo CPU mode being adjusted throughout the test. I waited for the values to stabilize in each test and cycle twice to and from Turbo Boost. I did the test once through counting results from when I started FurMark, but I realized when reviewing the video that the level that the GPU was at (whether I had just finished a run or left it idle for awhile for example) would affect the results greatly. So I did not count initial spikes while the GPU was below 80*C (and hence at a low fan power mode), but waited until any Turbo Boost activity was stable for at least 5 seconds minimum.

120W PSU rated at 87%+ Efficiency - 90W PSU rated at 84%+ Efficiency

120W Regular CPU Speed + Full brightness + Stock GPU Clocks: Avg. 113W, Peak 117W (caused by GPU Fan), Never any Turbo Boost

90W Regular CPU Speed + Full brightness + Stock GPU Clocks: Never any Turbo Boost, @ 108W Battery Mode to AC Mode pop ups keep cycling until Brightness is turned down

120W Regular CPU Speed + Lowest brightness + Stock GPU Clocks: Avg 110W, Peak 114W, Never any Turbo Boost

90W Regular CPU Speed + Lowest brightness + Stock GPU Clocks: Avg 99W, Never any Turbo Boost (CPU Core 0 was steady at 57*C and GPU steady at 87*C)

120W Regular CPU Speed + Full brightness + 400/1500/1000: 2.53GHz @ 118W max momentary peak down to 101W when Turbo Boost de-activated itself (Confirmed 115W stable for 5+ secs, CPU Core 0 temp 64*C AND GPU temp 85*C @ deactivation)

90W Regular CPU Speed + Full brightness + 400/1500/1000: 2.53GHz @ 109W max momentary peak down to 93W when Turbo Boost deactivated itself (Confirmed 108W stable for 5+ secs - CPU Core 0 temp 63*C and GPU temp 82*C @ deactivation)

120W Regular CPU Speed + Lowest brightness + 400/1500/1000: 2.53GHz @ 114W peak down to 99W when Turbo Boost de-activated itself (Confirmed 113W stable for 5+ secs - CPU Core 0 temp 65*C and GPU temp 85*C @ deactivation)

90W Regular CPU Speed + Lowest brightness + 400/1500/1000: 2.53GHz @ 104W peak down to 91W when Turbo Boost deactivated itself (Confirmed 104W stable for 5+ secs - CPU Core 0 temp 66*C and GPU temp 85*C @ deactivation)

120W Extreme Turbo CPU Speed + Full brightness + 400/1500/1000: Avg 105W, Peak 119W, Never any Turbo Boost (CPU Core 0 was steady at 58*C and GPU steady at 83*C) - (Bonus: Dim Screen 2.68 GHz @ 117W down to
103W when Turbo Boost de-activated itself - GPU 83*C, CPU Core 0 65*C at deactivation)

90W Extreme Turbo CPU Speed + Full brightness + 400/1500/1000: 95W Steady, Never any Turbo Boost (CPU Core 0 was steady at 58*C and GPU steady at 84*C)

120W Extreme Turbo CPU Speed + Lowest brightness + 400/1500/1000: 2.68GHz @ 115W momentary peak down to 97W when Turbo Boost deactivated itself (Confirmed 114W stable for 5+ secs - CPU Core 0 temp 65*C and GPU temp 86*C @ deactivation)

90W Extreme Turbo CPU Speed + Lowest brightness + 400/1500/1000: 2.68GHz @ 106W down to 91W when Turbo Boost deactivated itself (Confirmed 106W stable for 15+ secs - CPU Core 0 temp 67*C and GPU temp 83*C @ deactivation)

120W Extreme Turbo CPU Speed + Full brightness + Stock GPU Clocks: Avg 116W, Peak 120W, Never any Turbo Boost

90W Extreme Turbo CPU Speed + Full brightness + Stock GPU Clocks: Never any Turbo Boost, @ 109W Battery Mode to AC Mode pop ups keep cycling until Brightness is turned down

120W Extreme Turbo CPU Speed + Lowest brightness + Stock GPU Clocks: Avg 111W, Peak 114W, Never any Turbo Boost

90W Extreme Turbo CPU Speed + Lowest brightness + Stock GPU Clocks: Peak 108W, Never any Turbo Boost

Key Results:

90W Regular CPU Speed + Full brightness + 400/1500/1000: 2.53GHz @ 109W max momentary peak down to 93W when Turbo Boost deactivated itself (Confirmed 108W stable for 5+ secs - CPU Core 0 temp 63*C and GPU temp 82*C @ deactivation) -> For Sustained 108W Given 84% Minimum Efficiency, Minimum Actual System Wattage w/Turbo Boost Active: 90.7W

120W Regular CPU Speed + Full brightness + 400/1500/1000: 2.53GHz @ 118W max momentary peak down to 101W when Turbo Boost de-activated itself (Confirmed 115W stable for 5+ secs, CPU Core 0 temp 64*C AND GPU temp 85*C @ deactivation) -> For Sustained 115W Given 87% Minimum Efficiency, Minimum Actual System Wattage w/Turbo Boost Active: 100W

I spent many hours to write this post first by examining the PSUs and then researching the meanings, and ran and examined the tests very carefully and re-ran most of them 3 times to confirm the results. I'm not saying the BIOS isn't part of this - but I think that we also need to run a full series of tests on a 150W adapter. The combination Dell BIOS+Adapter upgrade may very well be what we need. I hope we can use these figures to compare with what we get on a 150W adapter doing these same things to confirm the Turbo Boost Status. Especially if it is possible to have Turbo Boost be at least relatively stable and active in FurMark with the GPU at Stock Clocks (as it was with the 400/1500/1000 underclock).

As was shown by my and AllurGroceries previous tests, you can push the 120W adapter to the expected 136W wall power draw but these tests did not do that because I did not overclock the GPU or use other USB devices - though if I had I suspect we could have reached the theoretical wall power draw limit (and possibly slightly beyond it) of the 120W adapter as we did the 90W adapter.

From the tests, CPU Turbo Boost when it disables itself, reduces total system power usage by ~15W, after which point it will reactivate itself again if slightly more than that power becomes available (fan speed reduction, screen dimming, etc...). It seems that if the system has enough power and is allowed to use it, then it should be able to support full GPU usage and full CPU Turbo Boost at the same time, or at least some level of CPU Turbo Boost and an overclocked GPU. Therefore, once and as long as we are within ~15W of the maximum power draw of the PSU, then the CPU throttles itself since there is not enough power to sustain the full mode. If it has determined that we need as much single core CPU power as we can get, as in the FurMark test, then it will keep trying to go to the full mode continually, rather than settling for a lower Turbo Boost that it can sustain with the power available.


 Posted - 2/24/2010 11:07:05 PM  - Show Profile  Edit Message

Site Administrator
2020 Contributions

Definitive Proof that the G51Jx GPU Overclock Performance is Power Restricted

All tests were with GPU Overclock 615/1950/1550 + Full Brightness + Extreme Turbo:
No USB Devices: 3DMark06 Score: 10342, SM2.0 Score 4356, HDR/SM3.0 Score 4470, CPU Score 3053
With rx1950 connected: 3DMark06 Score: 8702, SM2.0 Score 3372, HDR/SM3.0 Score: 3754, CPU Score: 3077
With rx1950 and Plasma Ball connected: 3DMark06 Score: 7190, SM2.0 Score 2511, HDR/SM3.0 Score: 3172, CPU Score: 3089
With Plasma Ball connected: 3DMark06 Score: 9309, SM2.0 Score 3655, HDR/SM3.0 Score: 4096, CPU Score: 3045

Throughout the tests, the GPU Tests averaged 5-20W or so lower power usage than without the USB devices connected. CPU Test Wattage were roughly equal though through all the tests and the scores matched as well.
I closed the WinMo connection window along with making sure that there was 0% CPU usage from it and everything else before running the tests.

The rx1950 has a power requirement of up to 5V @ 2A, and with Windows 7 the ActiveSync type software is included by default. To make sure it was not the data connection alone, I also tested with a Plasma Ball as well with 5V .3A marked power (5V .45A in my tests).

It appears as though the GPU needs a ceiling of power to go to full power and if you take the power available just below a certain level then performance will be affected. I say this because I noticed that the GPU when running in the 7190 test was actually using only 99W at the point where the Log lights up in the 'Firefly' GPU Test while when without any USB devices the power usage spiked there at 122-126W.

You can also notice that the stable GPU Overclock for 3DMark06 depends on the brightness (you can overclock slightly more if the brightness is set to lowest) - but this test shows a definite effect with the 120W adapter (as opposed to my previous 90W adapter tests) and just a slight difference in power of 2W when at that 615/1950/1550 overclock would result in a 10% performance difference for the GPU. 615/1950/1550 is the highest overclock that I found to be stable in 3DMark06 while at full brightness w/Extreme Turbo enabled with the 120W adapter.

The GPU temps never went above 85*C in any of these tests.

Therefore, imho, if someone can try a slightly higher overclock with a 150W adapter then that would be great :).


Last Edited By - PJPeter on 2/24/2010 11:11:50 PM
  This Topic is 16 Pages Long:    1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

Quick Reply

Not Registered? If So Click Here - It's Completely Free!


Your Username:
Your Password:
   Remember Your Username and Password.
Format Mode:
Format: BoldItalicizedUnderlineStrikethrough Align LeftCenteredAlign Right Horizontal Rule Insert Image Address Insert HyperlinkInsert Email Insert CodeInsert QuoteInsert List

 Include Your Profile Signature.
 Subscribe To This Thread.

rolex replica watches


Back to the Forum | Post Reply

  Bookmark and Share