• Announcements

    • Brian

      Introducing T|I Marketplace/Classifieds!   11/24/16

      After a long wait, Tech|Inferno now has a Marketplace/Classifieds section where all registered (promoted and above) members can participate in selling, trading and seeking out items. This service is free to all members to use as they wish. Currently all ads last for a period of 14 days before they expire for promoted users. Those users belonging to T|I Elite, Contributor, Developer and Moderator have access to 30 day ad periods that are featured   Some easy rules to remember: Do not post multiple ads of the same product. Do not advertise here if you are a retailer or business. You must contact Tech|Inferno for advertising space. Keep ads civil, do not try to lowball members. You may contact them in private via PM if you feel their price is not to your liking. Harassing members will result in an infraction. It is recommended you have Heatware/eBay feedback to establish trading credibility.    Tech|Inferno assumes no liability for any sales posted in the Classified section. Always use common sense when buying/selling with traders and ask for their heatware/ebay feedback!   Enjoy!
guilhrme

3dmark scores different for identical gpus.

13 posts in this topic

Hello,

I have an Alienware 18 (2013 model, A09 Bios), with two GTX 780m SLI (modded bios), Core i7 4930MX (Stock - Overclock enabled on the Bios Settings but in Default values), 32Gb Ram 1600mhz.

My friend has an very identical hardware, but his Alienware is a previous model, with a 3940MX Extreme Processor. The GPUs are identical, with modded bios too. Both systems with a 330W PSU.

Running 3dkmark, we realized that GPU scores has a significative difference.

After disabling Overclock option on the Bios, my scores grew a little more, but the difference between the systems continues.

Results: Result

Guys, do you think that can be my processor getting the power from the GPU's? If no, what could be?

Thank you.

1 person likes this

Share this post


Link to post
Share on other sites

Well, when one component grows too warm, it can trigger the thermal override throttling in other components. So it's certainly possible.

Share this post


Link to post
Share on other sites

I'm thinking you should both repaste your gpus and cpu then retest. This way it ensures it isn't a bad placement of the heatsink.

Share this post


Link to post
Share on other sites
I'm thinking you should both repaste your gpus and cpu then retest. This way it ensures it isn't a bad placement of the heatsink.

How a heatsink could affect the gpu performance since the temps are good? 50c Gpu 1, 51c Gpu 2 (idle)

Share this post


Link to post
Share on other sites

I think it's due to power limit.

Haswell used more power and that caused gpu to use less power.

Why dont you try single GPU test and see what happens? :)

Share this post


Link to post
Share on other sites
I think it's due to power limit.

Haswell used more power and that caused gpu to use less power.

Why dont you try single GPU test and see what happens? :)

The most curious thing is that even with a single GPU, my GPU score on 3dmark still less than the Alienware 18 (intel 3rd gen). Same mod, same driver, same clocks.

Did you ran this test?

Share this post


Link to post
Share on other sites

If 3dmark11 with a single GPU differs, power isn't the problem.

In that case, my best guess is that they might have different vbios.

Share this post


Link to post
Share on other sites

I only had m17x R4, m18x R2.

Recently I had a chance to test things on m17x R5 (AW17).

And found out that AW17 's GPU PCIE is not as good as her older sisters.

Even without CPU overclock and even with stock GPU clock, the graphic score of 780m or 880m in AW17 was lower.

Not only mine but every other AW17 had the same graphic score which is significantly lower than older sisters.

About 3~4% of graphic scores were lower in every AW17 compared to same clock GPU of m17x R4 or m18x R2.

(Even compared to old m17x R3 or m18x R1, haswell alienware's graphic score was lower)

So.. I guess this might be the problem of yours too.

I think AW18 has the same problem.

For better understanding, I am posting pics.

post-13289-144949985908_thumb.jpg

post-13289-14494998591036_thumb.jpg

post-13289-14494998591204_thumb.jpg

just look at the graphic score of each of three generation.

All stock GPU clock.

-----

I have recently posted this on Korean Alienware Forum. And someone said the cause might be the GPU or vbios difference.

So what I did was, I removed the same 780m from AW17 and put it on m17x R4.

Very good experiment to show how bad AW17's GPU PCIE is.

I believe there isn't a better one.

And the results are that AW17 has a 3-4% disadvantage in GPU score than m17x R4.

And the worst thing is if you compare a better GPU like 880m, the absolute GPU score difference gets bigger.

Which implies the score difference is not in "absolute value" but in percentage(3-4%).

Even worse thing is that AW17 has a worse slight graphic score than m17x R3 which has PCIE 2.0 while m17x R4 and AW17 are known to have PCIE3.0.

Haswell from the begining from now, they are giving us alot of problems.

Heats, throttle, ect.. and another one!

1 person likes this

Share this post


Link to post
Share on other sites

I think you are right because Alienware 18 seems to suffer exactly the same thing. Excellent testing and reporting Godfafa. Thanks for posting your findings.

Share this post


Link to post
Share on other sites
I think you are right because Alienware 18 seems to suffer exactly the same thing. Excellent testing and reporting Godfafa. Thanks for posting your findings.

One more thing to add.

I am sure It is not the CPU(i.e. Haswell) that causes this problem because graphic scores of 780m of clevo/msi/asus Haswell notebooks are similar with that of m17x R4 which is around 8100 in 3dmark11. Only AW17 780m shows 7700~7800.

So it must be caused by AW17 and AW18 mobo.

2 people like this

Share this post


Link to post
Share on other sites
One more thing to add.

I am sure It is not the CPU(i.e. Haswell) that causes this problem because graphic scores of 780m of clevo/msi/asus Haswell notebooks are similar with that of m17x R4 which is around 8100 in 3dmark11. Only AW17 780m shows 7700~7800.

So it must be caused by AW17 and AW18 mobo.

Thanks godfafa. Your comments and tests help me a lot. I'm very disappointed with AW18 and Alienware overall. I hope that on the next release Dell make things better.

Share this post


Link to post
Share on other sites
One more thing to add.

I am sure It is not the CPU(i.e. Haswell) that causes this problem because graphic scores of 780m of clevo/msi/asus Haswell notebooks are similar with that of m17x R4 which is around 8100 in 3dmark11. Only AW17 780m shows 7700~7800.

So it must be caused by AW17 and AW18 mobo.

Do you know whats different about the AW 18 that makes it suck so much more than the m18x or other gaming laptops?

Share this post


Link to post
Share on other sites
Do you know whats different about the AW 18 that makes it suck so much more than the m18x or other gaming laptops?

No idea why.

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!


Register a new account

Sign in

Already have an account? Sign in here.


Sign In Now

  • Similar Content

    • By lowteq
      Hello.
       
      I own an Alienware M17x-R3 and I'd like to upgrade the GPU.  The current GPU is an Nvidia 460M, and i've acquired and installed a 780M.  Unfortunately I'm running into a common problem, after I enable the IDT audio device in the BIOS Windows either goes to a blank or solid coloured screen right at the point where the Windows credential manager displays (the password prompt).  I believe I've followed all of the instructions I've read about on this and other forums, but I'm still having the problem.
       
      Here are my details
      Laptop:  M17x-R3
      Original GPU:  Nividia 460M
      New GPU:  780M
      OS:  Windows 7 x64 and Windows 10 x64, same symptom on both.
      New Graphics Driver Version:  Nvidia 364.51
      IDT Graphics Driver Version:  v.6330
      BIOS:  A12 Unlocked
       
      One concern I have is the 780M GPU I have is of unknown provenance, it came out of some other laptop, but I don't know the manufacturer.  I do believe it is working because with HD audio disabled it runs fine with no crashes.
       
      Here is exactly what I've done.
       
      With the laptop still in it's stock unmodified state I did the following.
      Installed the A12 Unlocked BIOS  Downloaded modded display drivers from here (http://downloads.eurocom.com/support/drivers/upgradeyourlaptop/Alienware/) I've tried the 364.51 drivers for Windows 7 and Windows 10.  I didn't install them at this point. Used the GURU 3D tool in safe mode to completely clean out the existing NVIDIA drivers (http://www.guru3d.com/files-details/display-driver-uninstaller-download.html) Rebooted into the BIOS and disabled the HD Audio, Disabled the Internal Graphics and set the primary display to PEG. I removed the old GPU, attached the existing heat sink and pipe to the new GPU using thermal paste and pads and installed all of it back in the laptop. Turned the laptop on and immediately went into the BIOS and confirmed the NVidia graphics were detected. Booted into Windows with digital signature verification for drivers disabled and installed the 364.51 modded driver I downloaded in step 2. Rebooted and did a few tests, display had the proper resolution and device manager indicated that I had a 780M installed. Rebooted went into the BIOS and enabled the High Definition audio, then immediately booted into safe mode. Installed IDT drivers version v.6330 (http://www.dell.com/support/home/uk/en/ukbsdt1/Drivers/DriversDetails?driverId=Y037W) in safe mode. Booted into normal mode and I get stopped at either a blank or solid coloured screen right when the credential manager shows  up (the password prompt)  
      I've tried a few other things including various combinations of versions of the Nvidia and IDT drivers, Windows 7 OEM image and Windows 10 scratch install.  Various orders of enabling the HD Audio and graphics.
       
      There are two things I haven't been able to try but which I saw suggested.
       
      Enable the Internal Graphics and HD Audio, remove the 780M GPU and boot using Internal Graphics, then do one or both of the following
      Install the IDT Driver at this point Disable Audio power management in the advanced tab of the IDT Control Panel Applet After doing those things disable Internal Graphics and HD Audio again, reinstall GPU, boot into Windows and Install Nvidia drivers, then reenable HD Audio.
       
      I've been unable to do these things because when I enable the Internal Graphics and remove the 780M I get no display at all and need to reset the BIOS to default using the Fn power one technique, after reinstalling the GPU.
       
      I'd appreciate some direction on how to proceed with troubleshooting.
       
       
    • By Mr. Fox
      For those that are not already aware of the issue, the folks at Futuremark seem to be struggling to keep a consistent product in the latest 3DMark benchmark. In particular, Fire Strike. Sometime around the release of Time Spy things started getting screwy with Fire Strike and now it seems with every Fire Strike GUI version update the effect is progressively decreasing benchmark scores, and specifically the physics portion of the benchmark.
       
      Kudos to @Papusan for noticing this months ago and asking me to have a look at it. He has been going back and forth with Futuremark about the problem and it seems they are either ignoring him or perhaps they do not view it as a high priority issue. Or, maybe because most people running Fire Strike are not observant enough to notice, care, or ask questions they feel they don't need to fix it.
       
      Some people might say you cannot compare results across benchmark software versions, but that shouldn't hold water here. There is a leaderboard and searchable database of results that basically every benching enthusiast and PC reviewer relies on, and if there is not a very high degree of consistency between GUI versions the results in their database will become irrelevant, as will their leaderboard. The search filter does not have a field to filter by GUI version, so we can expect the results from the database and leaderboard to be increasingly misleading, inaccurate and unreliable over time. This certainly is not a desirable thing for what is supposedly the current defacto standard in PC benchmarks.
       
      You will notice from the examples posted below that with each new version of Fire Strike the scores get lower and lower. These examples are consecutive runs on the same day, same machine, and identical CPU and GPU settings. The only thing that changes is Fire Strike benchmark results degrade with newer versions. We need Futuremark to understand and correct this.
       
      http://www.3dmark.com/compare/fs/11047304/fs/11047179/fs/11047154
       
       
      Here is a similar example from @Papusan: http://www.3dmark.com/compare/fs/11036017/fs/11035883
       
      If you agree this is a problem and want it to be fixed, please complain to Futuremark and let them know they need to put the brakes on and not do anything else with 3DMark until they have this mess under control. Gimmicky features are one thing, but inconsistent benchmark results makes 3DMark unreliable.
       
      If you would like to do your own testing to validate the issue before contacting Futuremark, older versions of 3DMark are available for download from the TechPowerUp.com web site. 
       
      In case you're not good at simple math, here is a visual aid to show what the fuss is about.
       

       
      Update 12/13/2016:
      We would like to acknowledge that a representative of Futuremark has responded promptly to this article and provided an email address for those interested in communicating with them about the issue. We appreciate the accountability and responsiveness. 
      Update 12/15/2016:
      We sincerely are grateful for Futuremark's responsiveness. I provided additional test results to Mr. Kokko to corroborate the findings of @Papusan and they have released an update that is expected to resolve the issue. See the message from James below for more details.

      View full article
    • By Mr. Fox
      For those that are not already aware of the issue, the folks at Futuremark seem to be struggling to keep a consistent product in the latest 3DMark benchmark. In particular, Fire Strike. Sometime around the release of Time Spy things started getting screwy with Fire Strike and now it seems with every Fire Strike GUI version update the effect is progressively decreasing benchmark scores, and specifically the physics portion of the benchmark.
       
      Kudos to @Papusan for noticing this months ago and asking me to have a look at it. He has been going back and forth with Futuremark about the problem and it seems they are either ignoring him or perhaps they do not view it as a high priority issue. Or, maybe because most people running Fire Strike are not observant enough to notice, care, or ask questions they feel they don't need to fix it.
       
      Some people might say you cannot compare results across benchmark software versions, but that shouldn't hold water here. There is a leaderboard and searchable database of results that basically every benching enthusiast and PC reviewer relies on, and if there is not a very high degree of consistency between GUI versions the results in their database will become irrelevant, as will their leaderboard. The search filter does not have a field to filter by GUI version, so we can expect the results from the database and leaderboard to be increasingly misleading, inaccurate and unreliable over time. This certainly is not a desirable thing for what is supposedly the current defacto standard in PC benchmarks.
       
      You will notice from the examples posted below that with each new version of Fire Strike the scores get lower and lower. These examples are consecutive runs on the same day, same machine, and identical CPU and GPU settings. The only thing that changes is Fire Strike benchmark results degrade with newer versions. We need Futuremark to understand and correct this.
       
      http://www.3dmark.com/compare/fs/11047304/fs/11047179/fs/11047154
       
       
      Here is a similar example from @Papusan: http://www.3dmark.com/compare/fs/11036017/fs/11035883
       
      If you agree this is a problem and want it to be fixed, please complain to Futuremark and let them know they need to put the brakes on and not do anything else with 3DMark until they have this mess under control. Gimmicky features are one thing, but inconsistent benchmark results makes 3DMark unreliable.
       
      If you would like to do your own testing to validate the issue before contacting Futuremark, older versions of 3DMark are available for download from the TechPowerUp.com web site. 
       
      In case you're not good at simple math, here is a visual aid to show what the fuss is about.
       

       
      Update 12/13/2016:
      We would like to acknowledge that a representative of Futuremark has responded promptly to this article and provided an email address for those interested in communicating with them about the issue. We appreciate the accountability and responsiveness. 
      Update 12/15/2016:
      We sincerely are grateful for Futuremark's responsiveness. I provided additional test results to Mr. Kokko to corroborate the findings of @Papusan and they have released an update that is expected to resolve the issue. See the message from James below for more details.
    • By Mr. Fox
      For those that are not already aware of the issue, the folks at Futuremark seem to be struggling to keep a consistent product in the latest 3DMark benchmark. In particular, Fire Strike. Sometime around the release of Time Spy things started getting screwy with Fire Strike and now it seems with every Fire Strike GUI update the effect is lowered benchmark scores, and specifically the physics portion of the benchmark.
       
      Kudos to @Papusan for noticing this months ago. He been going back and forth with Futuremark about the problem and it seems they are either ignoring him or don't care. Maybe because most people are not observant enough to notice or care.
       
      Some people might say you cannot compare results across benchmark software versions, but that shouldn't hold water here. There is a leaderboard and searchable database of results and if there is not a high degree of consistency between GUI versions the results in their database will become irreleant, as will their leaderboard. The search filter does not have a field to filter by GUI version, so we can expect the results from the database and leaderboard to be misleading, inaccurate and unreliable.
       
      You will notice from the examples posted below that with each new version of Fire Strike the scores get lower and lower. These examples are consecutive runs on the same day, same machine, and identical CPU and GPU settings. The only thing that changes is Fire Strike benchmark results. We need Futuremark to understand and correct this.
       
      http://www.3dmark.com/compare/fs/11047304/fs/11047179/fs/11047154
       
       
      Here is a similar example from @Papusan: http://www.3dmark.com/compare/fs/11036017/fs/11035883
       
      If you agree this is a problem and want it to be fixed, please complain to Futuremark and let them know they need to put the brakes on and not do anything else with 3DMark until they have this mess under control. Gimmicky features are one thing, but inconsistent benchmark results makes 3DMark unreliable.
       
       

      This post has been promoted to an article
    • By Doki
      Clevo X7200 for sale! good condition.
      with Windows 7
       
      config: price 590 USD
      CPU i7 960
      12GB RAM
      128GB SSD
      GTX 470M
       
      Barebone (cpu only) 460 USD
       
      without CPU 420USD
       
      ask for details