The 51%/98% GPU usage dilemma

Page 2 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

monstercameron

Diamond Member
Feb 12, 2013
3,818
1
0
I would have gone with the Hellcat.

Dodge sells a lot more v6 challengers than it does Hellcats (or before that, SRT8's) because people say "hey, it looks a lot like a Hellcat so it's cool to drive even if it's not that fast"
Thats closer to a 980ti vs titan scenario than what i was comparing.
 

poofyhairguy

Lifer
Nov 20, 2005
14,612
318
126
This is why AMD fails. Even after a presentation about their GPU launching it, they can't deliver a solid performance expectation.... Worst marketing department ever.

The truth is that AMD's problem isn't the marketing, it's the products.

Fiji was supposed to beat Nvidia's best but it didn't. The 290 blower was a product problem. Not having anything that matches a 1080 is a product problem.
 

Midwayman

Diamond Member
Jan 28, 2000
5,723
325
126
If the 480 was as fast as the 1080 in a single card or close they would have said that.

If it could outright be faster in crossfire they would have said that as well.

Instead you get this weird lower utilization thing implying that there is a lot more headroom. It kind points to them picking a pile of settings that severely limited the 1080 in some way. Think the whole tessellation issue previously. I'm going to go with something external to the GPU in this case though or else they would have just shown the crossfire setup much faster. CPU/memory/bus limits? Given the history my money is on super heavy async computer settings that are CPU bound.
 

TheELF

Diamond Member
Dec 22, 2012
3,991
744
126
Do you guys remember,once upon a time,where the ashes dev talked about the game and stated that it(the game) takes advantage of the number of cores,not the speed of cores?
Yup he was talking about the compute cores of GPUs, 2 480s have more cores then one 1080.
2x2304=4608 against the 3840 of the 1080.
And since all the game cares about is the number of cores and not the speed the cards work on low utilization hence the 51% (or maybe they did tweak it for drama reasons)

Remember though that this only works in benchmarks,further keep in mind that ashes was specifically build to showcase mantle and AMD hardware.

https://www.youtube.com/watch?feature=player_detailpage&v=9cvmDjVYSNk#t=63
Do remember that at around that time AMD tried to push the idea of every core being the same and started selling 10-12 "core" cpus (4CPU+6-8gpu cores) ashes even runs on "the newest APU" in the video.
 

Armsdealer

Member
May 10, 2016
181
9
36
I dont know man, halo products can be useful for sure...but there has to be a breaking point in the that logic.

Car analogy time!

Its like saying you want a sentra because GTRs (GOJIRA) are cool.

I think the analogy works with Ferrari. The Ferrari California is a very mediocre car in terms of actual performance (significantly worse than a 911 turbo and perhaps only a bit better than bmw m3...) But it's a Ferrari.
 

Stuka87

Diamond Member
Dec 10, 2010
6,240
2,559
136
The truth is that AMD's problem isn't the marketing, it's the products.

Fiji was supposed to beat Nvidia's best but it didn't. The 290 blower was a product problem. Not having anything that matches a 1080 is a product problem.

Polaris was NEVER intended to go against GP104. It was from day one a mainstream product. Just because AMD elected to launch its mainstream product before it high end product, does not mean it has a problem. High end cards are a small part of the market, AMD prefers to hit the largest part of the market first.
 

Irenicus

Member
Jul 10, 2008
94
0
0
If a single rx 480 is slightly slower then a 1080 then its faster than a 1070 and far cheaper. They could have just used a 1070.

If it's a decent bit slower it may be on par with a 1070 or even slightly slower. So why not mask it with a crossfire vs 1080 test. In a game with great crossfire scaling as well. Then take it back a bit with frame rate targeting to get the efficiency win too.


Amd is being vague though as usual. Nvidia's launch was better. By far. Even with the founder's edition pricing mess.

My guess is that a 1070 is still substantially faster than a 480, remember pascal, while not having true support for concurrent graphics + compute workloads, does seem to have much better preemption, and that allows them to not be as handicapped compared to maxwell in some of these games. Still not as much of a gain as async support was for amd, but not as much as a handicap.

This is just a guess on my part though, we got the sort of mysterious choice performance charts that make you worried, like when amd released fury x benchmarks, later found to be with questionable settings.
 

Sonikku

Lifer
Jun 23, 2005
15,752
4,562
136
The game was CPU bound at that point so it couldn't send the GPU more work, so it was going unutilized.

They didn't want to show single 480 vs 1080 since it would be (slightly?) slower, and people would ignore the huge $200-250 vs $700 price difference.

Kinda like 290 vs 780 Ti all over again, except even bigger price difference.

What leaves me scratching my head is the settings used. The Nvidia screen looked like it had more effects turned on. If the settings are not identical then it isn't an Apples to Apples comparison.
 

poofyhairguy

Lifer
Nov 20, 2005
14,612
318
126
Polaris was NEVER intended to go against GP104. It was from day one a mainstream product. Just because AMD elected to launch its mainstream product before it high end product, does not mean it has a problem. High end cards are a small part of the market, AMD prefers to hit the largest part of the market first.
I will agree that just because AMD doesn't put out the fastest card first isn't a problem in and of itself. The problem is if AMD never catches up- ie it doesn't launch a card that can beat a 1080 until after the 1080 ti/Titan X2 is released, or it never launches a card that beats a 1080 ti until after the 1180 comes.

Without a true halo product AMD will not be able to undo the marketshare position Maxwell put them in. It isn't enough to beat Nvidia at its chosen price segment, heck AMD has dominated value in the sub $300 segment for years now only for a crappy GTX 960 to eat their lunch. Eventually AMD needs to dominate period if they are gonna change consumer perception which is needed to sell more of those $200 cards.

Or maybe AMD has given up, I don't know. If I was them I would be more focused on making sure Sony has AMD in their next console, or that Apple has AMD in the next MacBook Pro, than trying to convince all those GTX 970 owners to wait for Volta instead of buying a 1070 as soon as they can. At some point it you have to be practical.
 

exar333

Diamond Member
Feb 7, 2004
8,518
8
91
I would have gone with the Hellcat.

Dodge sells a lot more v6 challengers than it does Hellcats (or before that, SRT8's) because people say "hey, it looks a lot like a Hellcat so it's cool to drive even if it's not that fast"

Yup. Then just slap that 'Type R' sticker on it for an extra 50hp.
 

sirmo

Golden Member
Oct 10, 2011
1,014
391
136
A scenario where AMD is still relevant to most GPU buyers despite not having a competitor for the entire top part of Nvidia's lineup for most (all?) of 2016.

The sad truth about technology is that halo products influence the purchase of non-halo products.
This is a myth, I wish people would stop repeating it. Halo product is pointless. 4870 and 4850 killed and there were no halo products. Also right now we do have the single fastest card on the planet in AMD's court the Radeon Pro Duo.
 

poofyhairguy

Lifer
Nov 20, 2005
14,612
318
126
This is a myth, I wish people would stop repeating it. Halo product is pointless.

Someone better tell Nvidia then, as we will probably see a 4th gen Titan within a year. I mean how else do you explain the success of the GTX 960 without accepting that it lived in the 980, Titan X and 980 ti's halo? The 280x was faster, the 380x is faster. I can buy the power savings argument for the 750 ti that doesn't need a 6 pin, but the 960 was a clear example of Nvidia's marketing winning when their technology lost.

Also right now we do have the single fastest card on the planet in AMD's court the Radeon Pro Duo.

Even AMD admits that is a VR card. Crossfire might as well not exist if you are a zero day gamer, which is the kind of gamer that keeps the whole wheel spinning.
 

Cookie Monster

Diamond Member
May 7, 2005
5,161
32
86
This is a myth, I wish people would stop repeating it. Halo product is pointless. 4870 and 4850 killed and there were no halo products. Also right now we do have the single fastest card on the planet in AMD's court the Radeon Pro Duo.

4870X2 was AMDs halo product back in those days.

And that single fastest card you mentioned is not only redundant (the GTX1080 comes hairy close for much less power and less then half the price), no one in their right mind would buy one. It simply doesn't belong in the discussion.

Halo products aren't pointless. Imagine Audi without its halo products
(A7 and up wards). You'd instantly lose the feel of premium behind the brand name especially when compared to its competitors. This applies to all type of products. For sure it may not affect a group of consumers, but for the majority it does and is why you see such products exist.
 

Glo.

Diamond Member
Apr 25, 2015
5,761
4,666
136
Two reasons why AMD GPUs may have been on 50% load.

Culling and Power Gating unused parts. Also if the GPUs were at 50% load, that means that they were not using more than 75W of power, each, IF we assume that power draw equals TDP, which I am pretty sure with this card is not true.

Polaris 10 may bring more surprises than we think...

But why the bloody hell being so silent?
 

Vesku

Diamond Member
Aug 25, 2005
3,743
28
86
Someone better tell Nvidia then, as we will probably see a 4th gen Titan within a year. I mean how else do you explain the success of the GTX 960 without accepting that it lived in the 980, Titan X and 980 ti's halo? The 280x was faster, the 380x is faster. I can buy the power savings argument for the 750 ti that doesn't need a 6 pin, but the 960 was a clear example of Nvidia's marketing winning when their technology lost.



Even AMD admits that is a VR card. Crossfire might as well not exist if you are a zero day gamer, which is the kind of gamer that keeps the whole wheel spinning.

IMO 960 sells well because reviewers and youtubers followed very tightly to Nvidia's guidance. Also the power efficiency is king internet campaign that has had a sudden reduction in intensity.
 

VulgarDisplay

Diamond Member
Apr 3, 2009
6,193
2
76
Two reasons why AMD GPUs may have been on 50% load.

Culling and Power Gating unused parts. Also if the GPUs were at 50% load, that means that they were not using more than 75W of power, each, IF we assume that power draw equals TDP, which I am pretty sure with this card is not true.

Polaris 10 may bring more surprises than we think...

But why the bloody hell being so silent?
They seriously need to clarify that slide. It was the worst part of the presentation because it made no sense without any context.

Sent from my XT1575 using Tapatalk
 

SelenaGomez

Member
May 30, 2016
92
3
11
ITs funny when they try to compare the P10 against the 970 and 980 when those cards were released 1.5 years ago lol. Its like Nvidia comparing the 1070 against the 7970.
 

SelenaGomez

Member
May 30, 2016
92
3
11
They seriously need to clarify that slide. It was the worst part of the presentation because it made no sense without any context.

Sent from my XT1575 using Tapatalk

They are being silent becuase they know the card isnt that good and its only selling point is that it is $200. I mean they are comparing the card to cards that are 1.5 years old. How lame.
 

exar333

Diamond Member
Feb 7, 2004
8,518
8
91
They are being silent becuase they know the card isnt that good and its only selling point is that it is $200. I mean they are comparing the card to cards that are 1.5 years old. How lame.

NV did the same thing with the 1080. The slides mostly referenced the 980 for performance and price comparisons. Obviously each company will use a comparison to their best advantage. I think it is actually telling that both companies compared against the same GPU (980) in many cases.

NV feels confident enough to compare against themselves (think Apple) because many of their customers are looking for an upgrade. AMD, on the other hand, is trying to woo customers away from buying NV. Both are pretty valid strategies as NV is the market share leader right now.
 
sale-70-410-exam    | Exam-200-125-pdf    | we-sale-70-410-exam    | hot-sale-70-410-exam    | Latest-exam-700-603-Dumps    | Dumps-98-363-exams-date    | Certs-200-125-date    | Dumps-300-075-exams-date    | hot-sale-book-C8010-726-book    | Hot-Sale-200-310-Exam    | Exam-Description-200-310-dumps?    | hot-sale-book-200-125-book    | Latest-Updated-300-209-Exam    | Dumps-210-260-exams-date    | Download-200-125-Exam-PDF    | Exam-Description-300-101-dumps    | Certs-300-101-date    | Hot-Sale-300-075-Exam    | Latest-exam-200-125-Dumps    | Exam-Description-200-125-dumps    | Latest-Updated-300-075-Exam    | hot-sale-book-210-260-book    | Dumps-200-901-exams-date    | Certs-200-901-date    | Latest-exam-1Z0-062-Dumps    | Hot-Sale-1Z0-062-Exam    | Certs-CSSLP-date    | 100%-Pass-70-383-Exams    | Latest-JN0-360-real-exam-questions    | 100%-Pass-4A0-100-Real-Exam-Questions    | Dumps-300-135-exams-date    | Passed-200-105-Tech-Exams    | Latest-Updated-200-310-Exam    | Download-300-070-Exam-PDF    | Hot-Sale-JN0-360-Exam    | 100%-Pass-JN0-360-Exams    | 100%-Pass-JN0-360-Real-Exam-Questions    | Dumps-JN0-360-exams-date    | Exam-Description-1Z0-876-dumps    | Latest-exam-1Z0-876-Dumps    | Dumps-HPE0-Y53-exams-date    | 2017-Latest-HPE0-Y53-Exam    | 100%-Pass-HPE0-Y53-Real-Exam-Questions    | Pass-4A0-100-Exam    | Latest-4A0-100-Questions    | Dumps-98-365-exams-date    | 2017-Latest-98-365-Exam    | 100%-Pass-VCS-254-Exams    | 2017-Latest-VCS-273-Exam    | Dumps-200-355-exams-date    | 2017-Latest-300-320-Exam    | Pass-300-101-Exam    | 100%-Pass-300-115-Exams    |
http://www.portvapes.co.uk/    | http://www.portvapes.co.uk/    |