AMD launches Radeon Pro DuoWorld’s Fastest Graphics Card With 16 TFLOPs

Page 6 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

railven

Diamond Member
Mar 25, 2010
6,604
561
126
they used a msi 390x which has the highest tdp of all 390x. doesnt mean much.

You could make the same argument for the 980/980 Ti. Regardless to that, I'm more interested to see what performance gains come with that huge power spike.
 

pj-

Senior member
May 5, 2015
483
251
136
they used a msi 390x which has the highest tdp of all 390x. doesnt mean much.

You don't think a 50% increase in power consumption means much? Take your favorite 390x's power consumption and add 40-50%, that's what it will be doing with async compute. Either that or they will hit power limits which will also limit performance.
 

pj-

Senior member
May 5, 2015
483
251
136
You could make the same argument for the 980/980 Ti. Regardless to that, I'm more interested to see what performance gains come with that huge power spike.

"While the Radeon R9 Fury X adheres to its power limit, the Radeon R9 390X under DirectX 12 blows through the roof compared to DirectX 11. Its 100W-higher power consumption with asynchronous compute activated may be okay when you consider the 20 percent performance increase, but it's just marginally less with this option deactivated, despite significantly lower framerates."
 

The Alias

Senior member
Aug 22, 2012
646
58
91
You don't think a 50% increase in power consumption means much? Take your favorite 390x's power consumption and add 40-50%, that's what it will be doing with async compute. Either that or they will hit power limits which will also limit performance.
Orrrr msi has their voltage too high. As logic would suggest. I figured power usage would go up with async usage because more computations more power usage, however the using the msi model with known crazy high power usage had to have some purpose.
 

AtenRa

Lifer
Feb 2, 2009
14,003
3,361
136
Simple, in order for the 390X to reach this performance level (32% faster than 980 and within 1-2% of 980Ti) it needs to use that much power. This is just a 430mm2 die that has the same performance as the 600mm2 GM200.

On the other hand the Fury X needs less power to even have higher performance.
 

tential

Diamond Member
May 13, 2008
7,348
642
121
One of our missing pro-AMD historical posters. of which I completely agreed with him. This card is beyond late. And now with that price tag, it's like AMD doesn't want to sell their products. I say give it two months or less before it gets a price cut.

(Although I might be off by citing November, was possibly December).
I'm done with Fiji. You can release this at 500. I'm not touching Fiji.
 

USER8000

Golden Member
Jun 23, 2012
1,542
780
136
Love the name, the design (if I ever owned one I'd rock it with the shroud off), but that price. Woof.

I recall few posters saying @$1000 it's DoA and this was back in like November. Isn't this a paper launch with actual sales not for another 1-2 months?

EDIT:

MOther of God, how have I not seen this before? Is this graph relatively new?



EDIT #2:
Found the THG article this is from, but without DX11 performance number, the power consumption means jack to me.

I thought Hawaii was a pig, DX12 Hawaii is a freaking boar!

The MSI R9 390X has very high power consumption:
https://tpucdn.com/reviews/MSI/R9_390X_Gaming/images/power_average.gif



It consumes 86W more than a reference R9 290X.

Look at the tests they did with the R9 390 cards:

https://tpucdn.com/reviews/Powercolor/R9_390_PCS_Plus/images/power_average.gif



https://tpucdn.com/reviews/Sapphire/R9_390_Nitro/images/power_peak.png



Look at the difference between the R9 290 and R9 290X cards too.

It seems the MSI R9 390X is a poor design.

If you look at some of the R9 390 cards tested by Techpowerup,they are still less than an R9 290X in power consumption.

A 113W difference between an R9 390X and an R9 390 - that is almost a 50% increase in power consumption!!
 
Last edited:

railven

Diamond Member
Mar 25, 2010
6,604
561
126
"While the Radeon R9 Fury X adheres to its power limit, the Radeon R9 390X under DirectX 12 blows through the roof compared to DirectX 11. Its 100W-higher power consumption with asynchronous compute activated may be okay when you consider the 20 percent performance increase, but it's just marginally less with this option deactivated, despite significantly lower framerates."

Ah, thanks for that. What I get for not reading the actual articles and just looking for DX11 numbers.

Definitely not worth flipping the DX12 switch for that configuration.
 

USER8000

Golden Member
Jun 23, 2012
1,542
780
136
Ah, thanks for that. What I get for not reading the actual articles and just looking for DX11 numbers.

Definitely not worth flipping the DX12 switch for that configuration.

The MSI R9 390X consumes 113W more,or nearly a 50% increase,over a Powercolor R9 390 and that is in a DX11 game.

This has been known since last year.

It is highly overvolted and anytime the card is put under any heavy load(like Metro Last Light) power consumption skyrockets.
 
Last edited:

tential

Diamond Member
May 13, 2008
7,348
642
121
The MSI R9 390X consumes 113W more,or nearly a 50% increase,over a Powercolor R9 390 and that is in a DX11 game.

This has been known since last year.

It is highly overvolted and anytime the card is put under any heavy load(like Metro Last Light) power consumption skyrockets.

Don't they also have an overvolted 970?
 

pj-

Senior member
May 5, 2015
483
251
136
The MSI R9 390X consumes 113W more,or nearly a 50% increase,over a Powercolor R9 390 and that is in a DX11 game.

This has been known since last year.

It is highly overvolted and anytime the card is put under any heavy load(like Metro Last Light) power consumption skyrockets.

Has anyone tested power consumption of a normal 390x in Ashes dx12 with async compute?
 

Erenhardt

Diamond Member
Dec 1, 2012
3,251
105
101
Has anyone tested power consumption of a normal 390x in Ashes dx12 with async compute?

If you have a hard time to grasp what everyone tries to tell you, do this simple test.

Set your bios to default, run some CPU benchmark and measure Power consumption.
Go to bios, and downlock your CPU to half the speed it runs on default. Change the voltages to 0.15V more than your default voltage.
Run the same CPU benchmark and measure power consumption.
Results?
On default CPU had higher score in benchmark than when it was downclocked and consumed less power than the downclocked one. The same CPU was faster and used less power. Magic!
 

pj-

Senior member
May 5, 2015
483
251
136
If you have a hard time to grasp what everyone tries to tell you, do this simple test.

Set your bios to default, run some CPU benchmark and measure Power consumption.
Go to bios, and downlock your CPU to half the speed it runs on default. Change the voltages to 0.15V more than your default voltage.
Run the same CPU benchmark and measure power consumption.
Results?
On default CPU had higher score in benchmark than when it was downclocked and consumed less power than the downclocked one. The same CPU was faster and used less power. Magic!

So is that a 'no' to my question?
 

Blue_Max

Diamond Member
Jul 7, 2011
4,223
153
106
JIBBERS CRABST!! Look at the size of that power connector!

I'll bet it's got a ray-shielded thermal exhaust port!
 

Raduque

Lifer
Aug 22, 2004
13,140
138
106
Do you put the little stickers on your case? Cute.


+1 :thumbsup:

Never understood the hate for case stickers. My old laptop, I left the nVidia/Intel stickers and added a Samsung SSD one. This laptop, I took off the energy star, g-sync and HDMI stickers, but left the gtx970 and Intel i7 ones (pny SSD didn't come with a sticker ). My last desktop I had Gigabyte, Seasonic, Intel and AMD stickers on.
 

KaRLiToS

Golden Member
Jul 30, 2010
1,918
11
81
Yes, I said that the Lepa and Enermax would run any load... Oh wait?

I had two Lepas out of 3 that failed. The last one was shutting off when pulling 900-1000watts. I RMAed it and sold it. I will never touch that PSU again. It is also poorly rated on newegg, and I would agree with those rating.

Just bought an EVGA SuperNova 1600 T2. It is night and day with the Lepa. I don't need to scratch my had to try to balance the load accross each rail. The rail are also really poorly made on the Lepa / Enermax units.

 

tential

Diamond Member
May 13, 2008
7,348
642
121
Never understood the hate for case stickers. My old laptop, I left the nVidia/Intel stickers and added a Samsung SSD one. This laptop, I took off the energy star, g-sync and HDMI stickers, but left the gtx970 and Intel i7 ones (pny SSD didn't come with a sticker ). My last desktop I had Gigabyte, Seasonic, Intel and AMD stickers on.
Because my pc is a pc not an advertisement for Intel amd and Nvidia etc.
 

KaRLiToS

Golden Member
Jul 30, 2010
1,918
11
81
Because my pc is a pc not an advertisement for Intel amd and Nvidia etc.

That is your opinion ( and also mine but I also respect others opinions).

Do you remove your car logos because you believe it is advertisement?
 

3DVagabond

Lifer
Aug 10, 2009
11,951
204
106
I had two Lepas out of 3 that failed. The last one was shutting off when pulling 900-1000watts. I RMAed it and sold it. I will never touch that PSU again. It is also poorly rated on newegg, and I would agree with those rating.

Just bought an EVGA SuperNova 1600 T2. It is night and day with the Lepa. I don't need to scratch my had to try to balance the load accross each rail. The rail are also really poorly made on the Lepa / Enermax units.


I can't understand why people are having so much difficulty understanding my personal preference for multi rail?

Since JonnyGuru.com has already been quoted.

One of the accessories bags contains these little doodads. What are they? Well, this power supply has a feature called "OCK," or "overclocking key." Basically, it's a way to disable the multi-rail overcurrent protection in the unit. You're not actually overclocking something. You can either use the little jumper on the left to permanently disable OCP, or plug in the expansion bay switch so that you can switch it on and off at will.

Folks, I'm telling you right now... this is a 1200 watt unit. Enough power to just about arc weld with. Try it with the multi-rail on first. Only use these doohickeys if the unit keeps shutting down on you no matter what cable combo you use.

See, this is why the "single 12V is better" thing is a myth. Sometimes, the bigger units of that ilk don't shut down when they should. They have too much current capacity. That's why my mining rigs are all running multi-rail when possible, even though they all have up to six video cards. If you plug things in right, and the power supply was designed correctly, you should never have an issue with multi-rail OCP shutting the unit down on you. But I've ranted enough about that before... I don't feel like getting into it again.

Now, I wasn't giving this as a recommendation to anyone or anything else. Just saying it so nobody would think I had a bias for single rail units. I can only assume people like to argue or because they personally had some issue they'd rather blame the hardware than take a chance it was something they might have done wrong. Dunno?

BTW Karl, None of this was directed at you. I simply made a statement that I preferred multi rail for the added OCP and I get the, "Multi rail is garbage" type of response out of nowhere. Then I get the anecdotal evidence presented to make a case against multi rail when none of it was solicited. I wasn't suggesting multi rail to anyone else where it required a debate over the benefits and shortcomings of each design.

Sorry, mate that you got dragged in.
 
Last edited:

railven

Diamond Member
Mar 25, 2010
6,604
561
126
That is your opinion ( and also mine but I also respect others opinions).

Do you remove your car logos because you believe it is advertisement?

Actually, when I bought my car new, I told the dealer I didn't want their emblem on it. Called me to pick up my new car, saw the dealer emblem, refused to accept it, showed them in writing my request that their sales person agreed upon.

Nulled transaction as they wouldn't honor what they agree'd to, bought same car from another dealer that would.

However, my case has little stickers :wub:
 

tential

Diamond Member
May 13, 2008
7,348
642
121
That is your opinion ( and also mine but I also respect others opinions).

Do you remove your car logos because you believe it is advertisement?

No just like I don't remove or paint over the shroud of my GPU.

I don't however add little stickers to other things to show that I have xyz car though.
 

thesmokingman

Platinum Member
May 6, 2010
2,302
231
106
I simply made a statement that I preferred multi rail for the added OCP and I get the, "Multi rail is garbage" type of response out of nowhere. Then I get the anecdotal evidence presented to make a case against multi rail when none of it was solicited. I wasn't suggesting multi rail to anyone else where it required a debate over the benefits and shortcomings of each design.

Sorry, mate that you got dragged in.


No, it started when you wrote that the 295X2 had a problem with multi-rail psus that it was a defect or some sort of assumption. When it was pointed out that that had more to do with not doing it right, that's when you went full tilt on multi-rail pwns choo, hyperbole yar...
 

3DVagabond

Lifer
Aug 10, 2009
11,951
204
106
No, it started when you wrote that the 295X2 had a problem with multi-rail psus that it was a defect or some sort of assumption. When it was pointed out that that had more to do with not doing it right, that's when you went full tilt on multi-rail pwns choo, hyperbole yar...

Um... no.

I was saying it was because it draws way over spec from the 8 pin connectors that it would trip some PSU's OCP.
 
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/    |