AMD Radeon RX Vega 64 8 GB

[Tsukiyomi91]

362W on Furmark with Standard Boost BIOS. Really AMD?? Is beating a one-year old GTX1080 non-Ti while increasing Vega's voltage limit for that?? NO. It's R9 3xx Series era of power consumption all over again. $500 asking price for that kind of performance? No thanks. Rather pick the GTX1080 non Ti for it's power efficiency.

 

[xkm1948]

Hey here is an idea. Can you do a clock to clock performance analysis between FuryX and Vega56/64 @W1zzard ? Really want to know if clocked the same where Vega's performance and power consumption level would be.

 

[Vya Domus]

View attachment 91091

Oh please , spare me the dank maymays , though this is TPU not Wccftech.

Not to mention it ain't got anything to do with what I said.

 

[medi01]

Fine... it trades punches with or barely beats a card released 444 days ago... all the while using a lot more power and being noisier.

Devil is in details:

Vega is a flop, not because of "444 days" etc (seriously, you are blaming underdog on verge of bancrupcy for not having yet another parallel project?) but that being that big (484 square mms vs 314 or 334, I was told) it rather lacks on performance front.
About 10% slower than similiarly sized nvidia card would be ok.

Power consumption is also ok-ish, in power saving mode (in which it is 10% slower than at max power mode)

 

[EarthDog]

Oh yeah....... do tell what you are trying to show there...

See your edit... What can you say to that... really... sure the power comes back to earth, yet still more than a 1080, when power savings is on and it loses 10% performance...

Yeah, those SLEW of DX12/Vulkan titles that are out or coming out saves it............. that is if production stopped from its competition. It takes looking at things through DX12 eyeglasses. So, let me go ahead and pay for this thing now, and HOPE that DX12/Vulkan titles start to saturate the market... meanwhile, 2+ years later, there will be new gen cards out.. LOL. Can't F with that infallible logic.

Hey, listen, I don't cut anyone slack... behind is behind after 444 days. This attempt is a bit disappointing unless the prices are able to stay low.

Oy...

 

[bug]

So overall ~20% performance increases from FuryX to Vega64 meltdown mode. Meh.

Vega64 is clocked 90% higher than Fury X, which makes me wonder what's under the hood.

 

[ERazer]

For a second runner up the pricing aint right needs to around $50 cheaper to make sense.

 

[bug]

Oh yeah....... do tell what you are trying to show there...

See your edit... What can you say to that... really... sure the power comes back to earth, yet still more than a 1080, when power savings is on and it loses 10% performance...

Yeah, those SLEW of DX12/Vulkan titles that are out or coming out saves it............. that is if production stopped from its competition. It takes looking at things through DX12 eyeglasses. So, let me go ahead and pay for this thing now, and HOPE that DX12/Vulkan titles start to saturate the market... meanwhile, 2+ years later, there will be new gen cards out.. LOL. Can't F with that infallible logic.

Hey, listen, I don't cut anyone slack... behind is behind after 444 days. This attempt is a bit disappointing unless the prices are able to stay low.

Oy...

And Nvidia keeps current prices. Which they have little reason to do after milking the market for over a year.

 

[B-Real]

I was a AMD fanboy, then I grew up. I realized Im a customer and I want whoever gives best performance, thus went to Nvidia this time and very happy with it.

Amazing to see people all over the net defending this card like there's no tomorrow, the power draw alone makes it a subpar card for the general customer, plus the price will go up due to mining.

To each their own...

There is price/performance ratio, you know. I think you should still grow up.

 

[medi01]

Vega64 is clocked 90% higher than Fury X, which makes me wonder what's under the hood.

About 60% higher.

 

[Captain_Tom]

It really goes to show that it depends on which games you play. I was seeing massive win after massive win, and then got confused by the "Performance Summary."



If you play Fallout 4, BF1, and DEUS EX - Vega is a completely different card compared to Watch Dogs 2 lol. By as much as 30%!

 

[Bjorn_Of_Iceland]

Let us wait for proper drivers guys. The same way we waited for Vega.


I mean, check it out! 5-7 fps faster than a GTX1080 in Deus Ex!

 

[Vya Domus]

and HOPE that DX12/Vulkan titles start to saturate the market...

Honestly with the way things are , I hope DX12 doesn't become the norm with the way it is used right now. At this point it is clear what's the deal with DX12. Many developers have stated that working with DX12 is a burden and sometimes it is difficult to even match what you can achieve with DX11 performance wise. Remedy Entertainment said they "feel at home with DX11" which says a lot , and we know how much of a clusterfuck Quantum Brake was on DX12. Most DX12 games we have either run the same as with DX11 or worse on both AMD and Nvidia.

MS needs to seriously consider stepping in and doing something about it.

 

[B-Real]

Hey here is an idea. Can you do a clock to clock performance analysis between FuryX and Vega56/64 @W1zzard ? Really want to know if clocked the same where Vega's performance and power consumption level would be.

Vega56 18% faster than FuryX with -8% power consumption.

 

[Manu_PT]

Is that power consumption really accurate? Wow

You dont want to live in a country where eletricty is expensive. Shocking numbers

 

[efikkan]

@W1zzard
Can you please explain your scale?
How can a terrible product like this get a 8.6 out of 10? And how bad does a product have to be to get something like a 5? What is the point of a scale 1-10 when there is barely any difference from a bad card like this vs. a good card like GTX 1080 (which got 9.1).

In my book, a 5 should mean a mediocre score, a completely OK decent product. Vega 64 fails to deliver in terms of performance, efficiency, noise, etc. It deserves a score of 3, 4 at most.

-----

Who in their right mind would buy Vega 56/64 for gaming? GTX 1070/1080 are clearly better options.

 

[DarkOCean]

Lol, these are the worst cards in AMD/ATI's history, even the 2900xt was better than this
It's time to find another hobby I guess.

 

[Manu_PT]

Hobby? Just switch to nvidia and keep doing your hobby!

 

[bug]

It really goes to show that it depends on which games you play. I was seeing massive win after massive win, and then got confused by the "Performance Summary."



If you play Fallout 4, BF1, and DEUS EX - Vega is a completely different card compared to Watch Dogs 2 lol. By as much as 30%!

Either you were reading a different review, or you have a very personal definition of "massive win".
Because on TPU, when it doesn't outright lose, the Vega 64 is within 10% of the GTX 1080 (save for Deus Ex and Doom, titles that have been always running better on AMD hardware).

 

[bug]

@W1zzard
Can you please explain your scale?
How can a terrible product like this get a 8.6 out of 10? And how bad does a product have to be to get something like a 5? What is the point of a scale 1-10 when there is barely any difference from a bad card like this vs. a good card like GTX 1080 (which got 9.1).

In my book, a 5 should mean a mediocre score, a completely OK decent product. Vega 64 fails to deliver in terms of performance, efficiency, noise, etc. It deserves a score of 3, 4 at most.

-----

Who in their right mind would buy Vega 56/64 for gaming? GTX 1070/1080 are clearly better options.

This has been asked in the past.

W1zzard cannot outright admit, has has hinted that if he scores lower, he's not going to get samples for much longer. I've stopped looking at that badge ages ago, even the pros and cons section can be a little tricky to read sometimes.

 

[efikkan]

I'm just waiting for people to start blaming "immature drivers" and "missing optimizations".

This is the worst top model released by AMD for years. Don't forget that Pascal is near the end of its life cycle, Vega would have to compete with Volta.

 

[Vya Domus]

Everyone is extremely polarizing throwing comprehensive arguments left and right such as : "this is a failure" , "worst thing ever" , "massive win" , "it's a flop" , "just switch to nvidia".

This has derailed into a fanboy showdown just as expect.

 

[the54thvoid]

It's hard to even make a comment without thinking it's going to annoy someone but I genuinely do not understand what RTG do all day to create so much hype about a card with so much technology that isn't better. Seriously. How? It's way faster than Fury X in clockspeed but it's not correspondingly faster in performance. Do RTG think of a million complex things to throw on a PCB and just hope it's all going to work? They need to refine, not increase the complexity.

No, sorry wait, refining a process is boring. People would rather see lots of new things that RTG can hype up that dont translate to real world gains. Has anyone noticed news about Navi's AI stuff is already leaking out?

RTG, already moving on and forgetting about Vega.

And before folk say it's the game developers fault for not implementing RTG's stuff, that's a wee bitty hypocritical because the same people argue Nvidia are bad for working with dev's. So again, for the company that have the Xbox and PS4 in their wings, why does their top range gfx card not do better when the consoles code for their chips?

 

[EarthDog]

@W1zzard
Can you please explain your scale?
How can a terrible product like this get a 8.6 out of 10? And how bad does a product have to be to get something like a 5? What is the point of a scale 1-10 when there is barely any difference from a bad card like this vs. a good card like GTX 1080 (which got 9.1).

In my book, a 5 should mean a mediocre score, a completely OK decent product. Vega 64 fails to deliver in terms of performance, efficiency, noise, etc. It deserves a score of 3, 4 at most.

-----

Who in their right mind would buy Vega 56/64 for gaming? GTX 1070/1080 are clearly better options.

Its subjective, sadly. Its why most sites have, and rightfully so, ditched the point system as there isn't anything quantitative attached to the values.

Everyone is extremely polarizing throwing comprehensive arguments left and right such as : "this is a failure" , "worst thing ever" , "massive win" , "it's a flop" , "just switch to nvidia".

This has derailed into a fanboy showdown just as expect.

You are't new here. Are you surprised?

 

[Steevo]

This has been asked in the past.

W1zzard cannot outright admit, has has hinted that if he scores lower, he's not going to get samples for much longer. I've stopped looking at that badge ages ago, even the pros and cons section can be a little tricky to read sometimes.

That has nothing to do with it. Read the review without your personal Bias. The frame pacing, pricing, and power consumption when chill and other features are turned on make it the *decent* card that it is.

It's not a huge win, it's more like the special Olympics winner that showed up to run at the regular Olympics.... juice box required.

AMD will probably fix half the issues with games, and improve performance by 5% while miners buy the card and they profit, some, cause HBM2 and the die isn't cheap.