• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

Pascal vs Maxwell **same clocks**

Tommy_Here

Tommy_Here

Tommy_Here

Associate
Joined
30 Dec 2013
Posts
2,238
Location
Liverpool
I don't know if that has been posted before but I've just stumbled across this video with a 980 Ti vs 1080 at the same clocks give or take a few mhz due to GPU boost, as you can see the 980 Ti is faster due to its cuda core advantage
what do you guys make of this? so with everyone mentioning drivers etc for the maxwell cards maxwell and pascal are so similar that I don't think drivers will matter as much as people make out with kepler and maxwell.
https://www.youtube.com/watch?v=nDaekpMBYUA
 
You can't compare differing cores with clock for clock. It just doesn't work like that. Compare a 1200Mhz Titan X with a 1200Mhz 680 and you will see a clear winner. A bit pointless in truth bud.

Edit:

Ohhh god, not him again!
 
Yea it's been known for a while that Pascal is pretty much Maxwell with a few tweaks. In theory Maxwell should not suffer as much as Kepler but we shall see.
 
Last edited:
Regardless of how inaccurate AMDTV is, Maxwell is proving to be years beyond anything AMD can come up with.

AMD needed 14nm to match the almost two year old 28nm GTX 970 in performance per watt.

Maxwell is one of the great milestones of GPU technology. Nvidia should be commended.
 
Mayh3M said:
Regardless of how inaccurate AMDTV is, Maxwell is proving to be years beyond anything AMD can come up with.

AMD needed 14nm to match the almost two year old 28nm GTX 970 in performance per watt.

Maxwell is one of the great milestones of GPU technology. Nvidia should be commended.
Click to expand...

Indeed. It has taken AMD nearly two years just to bring something out that matches the 970/980 in perf per watt.

It is slightly concerning in terms of what they can do to counter GP102 but hopefully vega has even further improved efficiency over POlaris and isn't just a scaled up Polaris. - HBM will obviously be used to alleviate the efficiency disadvantage as well (like with the fury range).
 
Mayh3M said:
Regardless of how inaccurate AMDTV is, Maxwell is proving to be years beyond anything AMD can come up with.

AMD needed 14nm to match the almost two year old 28nm GTX 970 in performance per watt.

Maxwell is one of the great milestones of GPU technology. Nvidia should be commended.
Click to expand...

You call him AMDTV but if you watched the Video at the end he basically says what you have about Maxwell. He believes that because Maxwell architecture is so efficient that's why Pascal came into existence as it was not on Nvidia's road map at one point.
 
TheRealDeal said:
You call him AMDTV but if you watched the Video at the end he basically says what you have about Maxwell. He believes that because Maxwell architecture is so efficient that's why Pascal came into existence as it was not on Nvidia's road map at one point.
Click to expand...

I call him AMDTV because I've read enough of his posts and watched enough of his videos to realise what his agenda is. You only need to take a glimpse at the comments sections of his videos to understand what it is he's pandering to.
 
Mayh3M said:
I call him AMDTV because I've read enough of his posts and watched enough of his videos to realise what his agenda is. You only need to take a glimpse at the comments sections of his videos to understand what it is he's pandering to.
Click to expand...

Well this video was about testing out Maxwell v Pascal to see if there is much difference. We all knew there was not going to be but he puts this down to Maxwell being so good on 28nm it made sense to shrink it with a few tweaks because it was so efficient. No matter who he bats for if any his video's are always pretty good and have reasoning behind his opinions.
 
Gregster said:
You can't compare differing cores with clock for clock. It just doesn't work like that. Compare a 1200Mhz Titan X with a 1200Mhz 680 and you will see a clear winner. A bit pointless in truth bud.

Edit:

Ohhh god, not him again!
Click to expand...

maybe that is the whole point, 680 to titan x is a clear winner, pascal 1080 to 980ti no winner ergo same performance per core.....
maxwell and pascal are awesome but i think the point is that for a huge price you're ending up with a re-branded small maxwell. stinks a bit of 290-390 to me but instead of getting drivers it get frequency.

although as he said its laziness comes from amd's incompetance this year(and last to an extent)
 
Gregster said:
Edit:

Ohhh god, not him again!
Click to expand...

Ya know... it wasn't long ago that we were all excited to see this guys videos and said things like, "I like this guys videos, he's not biased like the other guys" or "he makes a lot of sense" and so forth (many a post saying such things on this very forum).

I actually questioned him about his potential AMD bias and it seems not self-imposed, but his AMD coverage seems to have put him in a pothole regarding Nvidia. I.e. he didn't get a review card for the 1060 to benchmark himself (which he was not happy at all about), so he just used info from elsewhere. Nvidia just doesn't want anything to do with him cos he seems too pro-AMD.

I just wish he'd make a video addressing all his pre-release speculation and predictions and to talk about if he was on the ball or not. Would be interesting to see the guy recoup his own thoughts.
 
TheRealDeal said:
Well this video was about testing out Maxwell v Pascal to see if there is much difference. We all knew there was not going to be but he puts this down to Maxwell being so good on 28nm it made sense to shrink it with a few tweaks because it was so efficient. No matter who he bats for if any his video's are always pretty good and have reasoning behind his opinions.
Click to expand...

I think his logic is very basic. Most people could come to similar conclusions by themselves, but the truth is architecture is extremely complicated. Pascal seems to react very differently to Maxwell in DX12 for example. I think Pascal is similar to Maxwell in many ways, but to say they're basically the same is probably way of the mark.
 
mahius said:
I actually questioned him about his potential AMD bias and it seems not self-imposed, but his AMD coverage seems to have put him in a pothole regarding Nvidia. I.e. he didn't get a review card for the 1060 to benchmark himself (which he was not happy at all about), so he just used info from elsewhere. Nvidia just doesn't want anything to do with him cos he seems too pro-AMD.
Click to expand...

I remember someone asking him to make more Nvidia content, to which he replied along the lines of: What else is there to say about Pascal? We know it has terrible DX12 performance and non-existent A-Sync. Of course his minions were quick to upvote this.

Some of his forum posts elsewhere show he's not only heavily biased, but aggressively so. It's pretty obvious why he'll never receive anything from Nvidia. He barely hides his disdain for them.
 
Gregster said:
You can't compare differing cores with clock for clock. It just doesn't work like that. Compare a 1200Mhz Titan X with a 1200Mhz 680 and you will see a clear winner. A bit pointless in truth bud.

Edit:

Ohhh god, not him again!
Click to expand...

Every word sounds the same :p

But yeah not the best of videos to be honest.
 
Last edited:
One of the things with Pascal is that they've put back in some stuff stripped out for Maxwell which has little relevance to DX11 but will see more effect in later DX12+ titles hence Pascal can even be slower clock for clock when all else is equal.
 
To be honest the biggest issue with the video is the testing methodology. Nothing to do with comparing different cards, but the way he's getting the clockspeeds for the 1080 is just plain incorrect.

You can't just record the core clock once per second and expect that to be acceptable. A heavily power limited processor could be changing it's clockspeed hundreds of times per second, it's an incredibly fast operation and granularity of a second isn't anything like good enough. Maybe if you use a once a second measurement on a 10-20 minute benchmark run sure, but 30 seconds? No way is that good enough for a proper measurement IMO.
 
Mayh3M said:
Regardless of how inaccurate AMDTV is, Maxwell is proving to be years beyond anything AMD can come up with.

AMD needed 14nm to match the almost two year old 28nm GTX 970 in performance per watt.

Maxwell is one of the great milestones of GPU technology. Nvidia should be commended.
Click to expand...

Yes and no. Sadly in a lot of older games the 480 isn't any better then a 970 AMD should have really worked this out by now but when you enable full dx12 or Vulkan Polaris really pulls ahead as it can use all of its hardware to its fullest.
 
Tommy_Here said:
I don't know if that has been posted before but I've just stumbled across this video with a 980 Ti vs 1080 at the same clocks give or take a few mhz due to GPU boost, as you can see the 980 Ti is faster due to its cuda core advantage
what do you guys make of this? so with everyone mentioning drivers etc for the maxwell cards maxwell and pascal are so similar that I don't think drivers will matter as much as people make out with kepler and maxwell.
https://www.youtube.com/watch?v=nDaekpMBYUA
Click to expand...

If you drive a Ferrari and a Nissan Micra at 60Mph they are both as fast as each other. Save your money, don't buy a Ferrari, a micra is just as fast.


Really it is the most pointless comparison possible. Nvidia spent millions on critical path analysis and optimization in order to maximize the instructions per second, which is the only relevant indicator of performance. Critical path optimization will routinely lower IPC.
 
You must log in or register to reply here.
Back
Top Bottom