PASCAL-MXM & P-SERIES REFRESH

[D2ultima]

21 minutes ago, johnksss said:

 http://www.3dmark.com/fs/9595299

That is a 6700k, where do you get 6920hq from that?

 

I think you might want to go back and check first before jumping the gun my friend....

 

These are more closer to what you should be looking at.

http://www.3dmark.com/compare/fs/10011170/fs/10001580/fs/10059807

And like i have pointed out many times before. A single gpu physics score will always be higher unless the person benching does not know what they are doing by a very very long shot.

.....

http://www.3dmark.com/compare/fs/9954310/fs/10053831# this is the score comparison that @Mr. Fox posted with your score versus another user with a 6920HQ at 4GHz.

 

http://www.3dmark.com/fs/10053831 This is the score that I pulled from the above link just now, to look specifically at the Physics score (35.11fps; 11,059 score).

 

http://www.3dmark.com/fs/9595299 This is a run from @Spellbound, who I asked to provide one of her benches with her 6700K at 4GHz for me, so I could check her physics score (39.79fps; 12,534 score), as I felt that the above 6920HQ run was way too slow. Also, I noted that in his other benchmark (the comparison in Mr. Fox's post to your score in Time Spy and the MSI score in Time Spy was from the same person) his CPU only went up to 3.6GHz (either he removed the overclock, or something else happened).

 

My original post, saying "something is wrong", is pointing out that the 6920HQ at 4GHz should have minimum cracked 12K, and not sat at 11k on the PHYSICS test.

 

You bashed me, and posted a single GPU benchmark with a 4.7GHz chip in response. That was pointless. Then you told me that single GPU gets higher Physics scores than SLI setups, which while it may be true, does not equate to 1500 points lost. And then you post benchmarks for single GPU where the highest CPU speed is 3.6GHz, which does not equate to the 4GHz that was supposedly on the MSI benchmark.

 

What exactly have I "misunderstood" now? Unless if I go and disable SLI on my notebook and run a single GPU firestrike right now, I'll end up with a solid 1.5k increase in score?

[johnksss]

7 minutes ago, D2ultima said:

.....

http://www.3dmark.com/compare/fs/9954310/fs/10053831# this is the score comparison that @Mr. Fox posted with your score versus another user with a 6920HQ at 4GHz.

 

http://www.3dmark.com/fs/10053831 This is the score that I pulled from the above link just now, to look specifically at the Physics score (35.11fps; 11,059 score).

 

http://www.3dmark.com/fs/9595299 This is a run from @Spellbound, who I asked to provide one of her benches with her 6700K at 4GHz for me, so I could check her physics score (39.79fps; 12,534 score), as I felt that the above 6920HQ run was way too slow. Also, I noted that in his other benchmark (the comparison in Mr. Fox's post to your score in Time Spy and the MSI score in Time Spy was from the same person) his CPU only went up to 3.6GHz (either he removed the overclock, or something else happened).

 

My original post, saying "something is wrong", is pointing out that the 6920HQ at 4GHz should have minimum cracked 12K, and not sat at 11k on the PHYSICS test.

 

You bashed me, and posted a single GPU benchmark with a 4.7GHz chip in response. That was pointless. Then you told me that single GPU gets higher Physics scores than SLI setups, which while it may be true, does not equate to 1500 points lost. And then you post benchmarks for single GPU where the highest CPU speed is 3.6GHz, which does not equate to the 4GHz that was supposedly on the MSI benchmark.

 

What exactly have I "misunderstood" now? Unless if I go and disable SLI on my notebook and run a single GPU firestrike right now, I'll end up with a solid 1.5k increase in score?

I think you should go and read my edit's...:)

Edit: Yep. Why don't you go disable sli and find out for yourself. I'll wait right here.

Unless you have a 6920hq/980N setup? Then I already know that answer as well.

Edited September 7, 2016 by johnksss

[D2ultima]

2 minutes ago, johnksss said:

I think you should go and read my edit's...:)

Your own scores show that 1080N does the same thing.

http://www.3dmark.com/fs/9988160 single GPU from you, 4788MHz, 14,790 score.

http://www.3dmark.com/fs/9954310 dual GPU from you. 4790MHz, 15085 score.

[johnksss]

8 minutes ago, D2ultima said:

Your own scores show that 1080N does the same thing.

http://www.3dmark.com/fs/9988160 single GPU from you, 4788MHz, 14,790 score.

http://www.3dmark.com/fs/9954310 dual GPU from you. 4790MHz, 15085 score.

Yep, because of an os slow down i can't seem to get straighten out yet.

Im actually 500+ points behind from what pap was saying.

http://www.3dmark.com/compare/fs/3489139/fs/7284874/fs/6599319

http://www.3dmark.com/compare/fs/3489139/fs/7284874/fs/6599319/fs/6040190

Edited September 7, 2016 by johnksss

[D2ultima]

1 minute ago, johnksss said:

http://www.3dmark.com/compare/fs/3489139/fs/7284874/fs/6599319/fs/6040190

The fourth one was at 3.9GHz, though. What I can do at some point is run a firestrike, disable SLI, run another, enable SLI, then run a final one, and see if scores are generally consistent across them for Physics or not. I will have to do it at 3.8GHz however, since 3.9GHz has been wonky on this particular 4800MQ ever since I've gotten it. Randomly shuts off the PC under stress if I use it. I figure I could toss more voltage at it and stabilize it, but I'd rather use what I know is stable.

 

This however will be later, as I'm about to play some Killing Floor 2 with a friend.

[Guest]

Well, one thing is for sure, with Windows 10 physics performance is all over the board. I have found it terribly inconsistent and difficult to draw good comparisons with. One can have a great deal of inconsistent results running the same machine with nothing that I have been able to pinpoint as the reason for erratic changes in physics performance. Windows 8.1 was that way to some extent also, but consistently worse. Windows 10 is just erratic... sometimes good, sometimes lousy.

[johnksss]

5 minutes ago, D2ultima said:

The fourth one was at 3.9GHz, though. What I can do at some point is run a firestrike, disable SLI, run another, enable SLI, then run a final one, and see if scores are generally consistent across them for Physics or not. I will have to do it at 3.8GHz however, since 3.9GHz has been wonky on this particular 4800MQ ever since I've gotten it. Randomly shuts off the PC under stress if I use it. I figure I could toss more voltage at it and stabilize it, but I'd rather use what I know is stable.

 

This however will be later, as I'm about to play some Killing Floor 2 with a friend.

My apologies though! I did not mean to offend

 

Side note

Good thing physics is on it's way out the door. 

Edited September 7, 2016 by johnksss

[Guest]

@D2ultima I don't know if you noticed, but Brother @Papusan has been complaining about Fire Strike physics score being goofed up for a while now. Since Time Spy was released they have been messed up. It's mostly a GPU test anyway and the physics scoring seems to have minimal effect on the overall score. The combined score has more of an effect, but even that is muted a little bit compared to 3DMark 11. It seems like more of a fluffy feel good benchmark that helps BGA turdbook owners not feel bad about owning something with a crappy CPU. If it took a powerful CPU to get a nice overall score the turdbook owners probably would not be running it. They would be avoiding it like most of them avoid 3DMark 11 and Vantage.

[johnksss]

Getting better @D2ultima

 

Higher Physics Score, but it still lost.

http://www.3dmark.com/compare/fs/10073122/fs/10073016

Edited September 7, 2016 by johnksss

[D2ultima]

27 minutes ago, Mr. Fox said:

@D2ultima I don't know if you noticed, but Brother @Papusan has been complaining about Fire Strike physics score being goofed up for a while now. Since Time Spy was released they have been messed up. It's mostly a GPU test anyway and the physics scoring seems to have minimal effect on the overall score. The combined score has more of an effect, but even that is muted a little bit compared to 3DMark 11. It seems like more of a fluffy feel good benchmark that helps BGA turdbook owners not feel bad about owning something with a crappy CPU. If it took a powerful CPU to get a nice overall score the turdbook owners probably would not be running it. They would be avoiding it like most of them avoid 3DMark 11 and Vantage.

Yes, I did see. I was wondering when they'd fix it.

[D2ultima]

57 minutes ago, johnksss said:

Getting better @D2ultima

 

Higher Physics Score, but it still lost.

http://www.3dmark.com/compare/fs/10073122/fs/10073016

Okay. I've run the three benches I said I would. I DID see a benefit on single GPU in the physics test. But it was very slight, as reflected in the results.

 

http://www.3dmark.com/compare/fs/10073292/fs/10073334/fs/10073378

 

As far as I can see, this is what should happen. The slight benefit, logically, is due to the fact that SLI in itself has a CPU overhead to render. The same should be for crossfire. It's probably a good thing people haven't figured out a way to use NVPI to specifically disable SLI via NVAPI for the physics test alone, or that would be a way to boost scores without actually turning SLI on/off inbetween tests (which would really be impossible I think).

 

As for why multi-GPU is doing better in Physics with Maxwell (and Pascal, since Pascal is essentially die-shrunk maxwell overclocked), you've got me stumped right there. I couldn't begin to guess. It makes no logical sense to me.

[Dr. AMK]

14 hours ago, Mr. Fox said:

I did a little benching with moderate overclocks connected via Team Viewer to Eurocom's campus with an unattended Sky X9E2 this afternoon. I was not able to control the fans and the CPU temps are totally out of control, so this was with whatever thermal paste they have and sitting on a flat surface with some pretty severe CPU thermal throttling. I could not push the CPU any further without direct access to the machine,

 

I don't know really how those companies are thinking, all respected companies I know are sending their products for reviewing, testing and bench-marking for free to the experts or for sales teams, even for the YouTube channels!? they must do the same with experts like our seniors members in this forum and I'm sure that they are seniors in other forums as well. very strange situation??!! This is the 1st time for me to see remote Overclocking 

Edited September 7, 2016 by Dr. AMK

[johnksss]

11 hours ago, D2ultima said:

Okay. I've run the three benches I said I would. I DID see a benefit on single GPU in the physics test. But it was very slight, as reflected in the results.

 

http://www.3dmark.com/compare/fs/10073292/fs/10073334/fs/10073378

 

As far as I can see, this is what should happen. The slight benefit, logically, is due to the fact that SLI in itself has a CPU overhead to render. The same should be for crossfire. It's probably a good thing people haven't figured out a way to use NVPI to specifically disable SLI via NVAPI for the physics test alone, or that would be a way to boost scores without actually turning SLI on/off inbetween tests (which would really be impossible I think).

 

As for why multi-GPU is doing better in Physics with Maxwell (and Pascal, since Pascal is essentially die-shrunk maxwell overclocked), you've got me stumped right there. I couldn't begin to guess. It makes no logical sense to me.

Physics is for the cpu. No GPU involvement.

 

I'll let you guys figure out the tech aspects of it. I'm just going to bench it and move on.

[Guest]

19 minutes ago, johnksss said:

Physics is for the cpu. No GPU involvement.

 

I'll let you guys figure out the tech aspects of it. I'm just going to bench it and move on.

I like that approach best. Pretty simple... The best score wins regardless of whether the benchmark works they way we think it should or not.

[johnksss]

Yep. 

 

[D2ultima]

2 hours ago, johnksss said:

Physics is for the cpu. No GPU involvement.

 

I'll let you guys figure out the tech aspects of it. I'm just going to bench it and move on.

Yes it is, but while the CPU does the calculations, it still has to feed to each GPU to display. I.E. each GPU ought to have a VERY slight load during the Physics test, and just feeding the extra GPU should theoretically take away an ever so slight amount of compute power from the physics test.

 

But this doesn't happen for Maxwell/Pascal and it makes no sense to me as to why.

[Guest]

32 minutes ago, D2ultima said:

Yes it is, but while the CPU does the calculations, it still has to feed to each GPU to display. I.E. each GPU ought to have a VERY slight load during the Physics test, and just feeding the extra GPU should theoretically take away an ever so slight amount of compute power from the physics test.

 

But this doesn't happen for Maxwell/Pascal and it makes no sense to me as to why.

I can't speak for Pascal (yet) but it does apply to Maxwell. I saw a slight decrease in physics scores with 980 SLI and 980M SLI versus single GPU on both the Sky X9 and Panther. And, that drop in physics scores was always slightly more of a hit with W8.X and W10 versus Windows 7.  I believe (can't prove) the issue is unnecessarily aggressive Windows power management tree-hugger filth that they made worse in the newer versions of Windows. Running in UEFI mode (versus Legacy BIOS) may have a slight degree of influence on it as well. Using ThrottleStop somewhat mitigates the diminution of physics performance running in SLI, which further contributes to the idea it is Windows-induced nonsense.  At any rate, since there is nothing we can do to fix it and we can only speculate as to why, probably best to just move on as @johnksss suggested. Interesting discussion topic though.

[D2ultima]

Eh but the minute I notice something being weird, I INSTANTLY want to find out why. How do you think I've learned so much already? xD

[Guest]

Just now, D2ultima said:

Eh but the minute I notice something being weird, I INSTANTLY want to find out why. How do you think I've learned so much already? xD

Oh, I understand that 100%. Overclock and bench more and game less and you'll probably find even more weird questions like this.

[Dr. AMK]

 

 

[D2ultima]

4 minutes ago, Mr. Fox said:

Oh, I understand that 100%. Overclock and bench more and game less and you'll probably find even more weird questions like this.

Why game less? Why not overclock and bench more AND continue gaming as much? =D

 

Besides, gaming has a lot of even weirder questions too. Also, anybody has any idea why the combined test for firestrike only scales 50% on SLI? At least for me. Single GPU was generally 11-14fps, and multi-GPU was 17-20fps. It's basically 50% boosting. But the graphics tests were over 95% scaling?

[Guest]

Just now, D2ultima said:

Why game less? Why not overclock and bench more AND continue gaming as much? =D

 

Besides, gaming has a lot of even weirder questions too. Also, anybody has any idea why the combined test for firestrike only scales 50% on SLI? At least for me. Single GPU was generally 11-14fps, and multi-GPU was 17-20fps. It's basically 50% boosting. But the graphics tests were over 95% scaling?

I guess that depends on whether you have that much spare time. If one does, sure... might as well. Since my spare time is limited I have to decide how to spend it and still get in 3 to 6 hours a day for sleep.

[D2ultima]

1 minute ago, Mr. Fox said:

I guess that depends on whether you have that much spare time. If one does, sure... might as well. Since my spare time is limited I have to decide how to spend it and still get in 3 to 6 hours a day for sleep.

What is this thing you say called "sleep"?

 

Jokes aside though, I am indeed currently unemployed, and thus I do have quite a bit of time.

 

Edit: Truthfully, the SINGLE reason I do not overclock or bench much is because it's simply too hot and humid and has extremely little airflow in my room. It's on average 90F per day and if it drops to 77F by midnight, I consider that very cool. It also has very little airflow, so when the area around me gets hot, it generally remains that way for a long time. Mythlogic was able to tell in an instant that I was in an extremely hot and humid environment the first time they got my notebook; simply six months after I had it. Anything beyond the 1006/6000 above is instantly non-useful for pretty much anything that isn't firestrike, just due to temperatures, unless it's a particularly cool night and I run benches at around 4am when it's about at its coldest.

 

If I get access to a cooler room in the day, preferably one with an A/C, I would pretty much keep daily driver overclocks in general. I've used some A/C rooms before, even large ones at university, and I was able to hold LONG periods of overclocking in even Crysis 3 without thermal throttling or crashing.

Edited September 7, 2016 by D2ultima

[Dr. AMK]

 

There is a serious problems with SLI to support some common games, this is really weird, can those problems be fixed by drivers or by BIOS or what?

There are a lot of benchmarks in this video if someone like to know.

Edited September 7, 2016 by Dr. AMK

[Dr. AMK]

I think we have to consider Delidding the 6700K to reduce the temp as nuch as we can. Do it by our self is risky for some of us, but now HIDevolution is giving this as an option beside the Silicon Lottery but with lower cost.

Find below 2 different ways for Delidding. 

Our seniors members please advice.

Regards.