AMD Radeon RX Vega 64 and 56 Reviews [*UPDATED* Aug 28]

[Elixer]

Lets just face reality: They aren't going to close the gap entirely with Navi vs. Volta. It just isn't going to happen. Polaris and Vega had fairly steep mountains to climb to catch Pascal, but instead of clawing back perf/w, the gap widened. So rather than saying "catch up to" Volta, or hoping that Navi "delivers," AMD just needs to move in the right direction.

You know, that is the exact same thing people were saying after bulldozer launched, AMD is dead, they can't compete, they are going bankrupt, and so on.
Reality showed that AMD can actually compete, once they have good leadership & direction, so, for all we know, Navi could be the comeback kid, like Ryzen was.
(And NOBODY will believe anything from AMD's crappy PR machine about Navi, until benches are done.)

AMD's Raven Ridge should be another winner for them, and you won't see Vega breaking power records there, I am betting we will see a pretty efficient version of Vega on die.
Interesting thing here is, obviously they can't fit Vega 10 in there, so, what parts are they "cutting out" to meet the space requirements?
Raven Ridge should be a hit with OEMs, and they will flock to it, and they will get a pretty big chunk of market share.

The thing about Vega 10(?) is, what reason(s) is Vega's performance so mediocre?
On paper, with the feature set Vega has, it should have been much better than what we are seeing (and heavily lead to believe), and with over 18 months in development, it is looking more and more like that there are hardware issues that must be bypassed by software (drivers) to fix.

The power demons were a result of being boxed into a corner, it couldn't compete against Nvidia's high end without cranking up the clocks as high as they could and they still couldn't solidly beat the 1080.

It also shipped with broken and/or disabled features in the drivers, which I think is related to having to rewrite driver routines to fix whatever is broken.

Will we ever find out what went wrong with Vega? Doubt it.
Will Navi be better? Can't be worse.
Raja will stay on until Navi ships, there is no way they will get rid of him before that, it wouldn't make any sense.

 

[PeterScott]

You know, that is the exact same thing people were saying after bulldozer launched, AMD is dead, they can't compete, they are going bankrupt, and so on.
Reality showed that AMD can actually compete, once they have good leadership & direction, so, for all we know, Navi could be the comeback kid, like Ryzen was.
(And NOBODY will believe anything from AMD's crappy PR machine about Navi, until benches are done.)

AMD's Raven Ridge should be another winner for them, and you won't see Vega breaking power records there, I am betting we will see a pretty efficient version of Vega on die.
Interesting thing here is, obviously they can't fit Vega 10 in there, so, what parts are they "cutting out" to meet the space requirements?
Raven Ridge should be a hit with OEMs, and they will flock to it, and they will get a pretty big chunk of market share.

Intel really isn't the fierce competitor that NVidia is. Relatively speaking, Intel was practically standing still waiting for AMD to get it's act together, NVidia is Not. So short of a miracle, I don't expect AMD to close the gap with NVidia.

AMD is trying to make a Jack of all trades (master of none) Video card, while NVidia makes HPC and Gaming focused cards, winning both markets.

 

[tential]

literally an hour of browsing online web boards and gaming communities would have revealed the nature of the demand that is "exceeding expectations..."

Hilariously enough, this is a 100% false statement.

If you browse online forums/gaming communities, you're not seeing a level of interest that would indicate that these cards would be sold out.

It's clearly low quantity. Which shouldn't even be a news story at this point. A GPU release has low quantity. isn't news. It's just normal.

 

[tential]

If Vega is indeed faster than nvidia without AA, that changes things a lot.

AA is crap in my opinion, i turned it off 99% of the time.

It's more like it looks like Vega's AA implementation is broken currently. I'm not sure why this subject, the MOST important to Vega's performance, is being ignored. We basically got review results that are half baked....

 

[dlerious]

The funny thing is people are complaining about the $100 increase yet i highly doubt most people were even looking to purchase Vega even when it was $100 less. The limited supplies are still selling out. Hardcore fanboys? Miners?

I tried. The MSI was the only card I saw at $499 when I checked. Clicked Auto-Notify but didn't get an email for it back in stock. Checked the page a few times after and the the card went to 'sold out' about 2 hours later and disappeared completely 2 hours after that. As for who's buying them, check Ebay and Amazon for all the scalpers (i.e. those buying at MSRP and adding $400 and up to the price).

 

[jpiniero]

It's clearly low quantity. Which shouldn't even be a news story at this point. A GPU release has low quantity. isn't news. It's just normal.

Miners are eating it up I imagine.

 

[Elixer]

Miners are eating it up I imagine.

Only Monero miners.
It is just people trying to flip them at a huge profit + the extremely low stock.

 

[tential]

I tried. The MSI was the only card I saw at $499 when I checked. Clicked Auto-Notify but didn't get an email for it back in stock. Checked the page a few times after and the the card went to 'sold out' about 2 hours later and disappeared completely 2 hours after that. As for who's buying them, check Ebay and Amazon for all the scalpers (i.e. those buying at MSRP and adding $400 and up to the price).

The only reason I checked Vega was to scalp at this point. $499 was the only MSI card. Me and @Phynaz lol'd at the "token" MSRP card there as a place holdler for the launch.

There's just no point in complaining about low quantities because we don't actually know just how low the stock is.

Finally... a quick economics lesson for all those who are confused as to "Who is buying Vega".
This is actually not hard to explain using economic theory.
In the picture below we have the demand curve.
You can see
Vega MSRP
Slightly Inflated MSRP
Price Vega would sell at if Markets functioned efficiently

At MSRP, Vega would sell a large quantity. However, that quantity isn't available. The inflated price, they'd still sell a decent quantity. But it's STILL not there. The actual quantity is lower. For the quantity supplied, it turns out the amount of buyers is few, but they'd all pay FAR higher prices because of their position on the demand curve. So actually they'd be willing to pay $1000+ (or whatever) and are getting Vega at what they consider to be a steal.

This has nothing to do with your opinion on Vega, it's just basic economics. Anyone is welcome to PM me for a further explanation.
If you'd like to learn on your own:
Economics in One Lesson: The Shortest and Surest Way to Understand Basic Economics
https://www.amazon.com/Economics-One-Lesson-Shortest-Understand/dp/0517548232

 

[Kenmitch]

Intel really isn't the fierce competitor that NVidia is. Relatively speaking, Intel was practically standing still waiting for AMD to get it's act together, NVidia is Not. So short of a miracle, I don't expect AMD to close the gap with NVidia.

AMD is trying to make a Jack of all trades (master of none) Video card, while NVidia makes HPC and Gaming focused cards, winning both markets.

The wise choice was diverting R&D funds to the cpu department. Parity or or superiority with Intel will bring way more $'s in the end. GPU should play 2nd fiddle to those who have x86 licenses.

Sadly when it comes to AMD's offerings way too many end users only want them too be competetive for better pricing on the other guys offerings.

The wall is out there. Not sure how far away it is but most likely Nvidia will hit it 1st.

 

[jpiniero]

Only Monero miners.
It is just people trying to flip them at a huge profit + the extremely low stock.

You really think so? It's the fastest Ethereum card out there and it's not like you can buy anything Polaris and 1070's go for at least $450 now.

 

[tviceman]

Vega closed some of the gap, it's just the equivalent of the RX 580 right now. Pushed to the limit.

In power saver 2nd BIOS, the card is at the heels of Pascal in efficiency and a good bit above 400 series Polaris (the none pushed to the limit ones).

20% more perf/watt than the best Polaris has to offer.

I see what you're saying, but you're also falling into this trap AMD wants you to dig yourself into. Do you think it's a smart move for people to spend $500-600 on the RX Vega 64 graphics card, put it in power saving mode where it ends up 10% slower and and still 25% less efficient than the likely-cheaper GTX 1080? And further, I'm sure if Nvidia wanted to put GTX 1080 into some kind of power saver / mobile mode on the desktop and drop power consumption down to 130 watts for a 10% loss in performance they could easily do that, thus making RX Vega's power saving mode still look as inefficient as it's balanced and turbo modes. But, you see, Nvidia doesn't have to waste time creating driver profiles that function in power saving modes because their cards don't draw an embarrassing amount of power anymore.

Vega is more than 50% larger in die size, has 60% more transistors, and consumes 120+ watts more when going toe-to-toe with the GTX 1080. The 390x was much closer in die size and transistors to it's immediate competitor (the GTX 980) while consuming similar or less power than RX Vega 64.

Vega is not a step in the right direction. Fiji was a huge step in the right direction (they were actually competitive with Fiji, OMG remember that?!). Polaris was a small stumble, Vega is lame duck. AMD can't keep throwing transistors and TDP at the problem; they're at the wall. If they don't heavily innovate they will be even worse off next round because I guarantee you Volta will not be some 5-10% update like Intel's CPU's were for years which allowed AMD to catch Intel in the CPU space.

 

[tviceman]

You know, that is the exact same thing people were saying after bulldozer launched, AMD is dead, they can't compete, they are going bankrupt, and so on.
Reality showed that AMD can actually compete, once they have good leadership & direction, so, for all we know, Navi could be the comeback kid, like Ryzen was.
(And NOBODY will believe anything from AMD's crappy PR machine about Navi, until benches are done.)

In the CPU space, AMD caught up to a company milking it's own tired design with very little gains and no additional cores for several years. Intel quit innovating and started milking. Nvidia has not rested on their laurels since Fermi 2.0. Kepler came out strong (even though it ended up falling behind), Maxwell came out stronger, and Pascal even more than Maxwell. Nvidia isn't milking their designs for 25-30% gains like Vega was over Fiji.

 

[zinfamous]

There's no such thing as excellent or terrible card just excellent and terrible prices and at $500 (actual non bait and switch msrp), Vega 56 is opposite of excellent

you may be right, because that's the prevailing rumor and the trend seems to be going that way with the 64, but there are no 56s for sale anywhere so we really don't know if that is the case.

I have 2 assumptions:

1: not "bait and switch," but the actual "Red packages" with the discount price are in short number and sell out very fast, and the cards slated for the full price are more abundant. Yes, sounds the same, but technically not.
or

2: Vega 56 is their target card for the gaming market. With the delays and delays based on, supposedly, unexpected silicon, there is a far greater number of 56 chips out there, and AMD will target these at the advertised price to project as best they can the typical value label that they live with. I'm betting that they want the 64 to stand in for the enthusiast market and will sell at whatever people are willing to pay, but they want 56 to be their workhorse Vega. At the very least, the profile seems to have the best potential for a positive perspective in the general consumer's mind, so kill it on price and value with the 56, and sell are far lesser number of 64 at whatever ridiculous price consumers are willing to pay.

or not

 

[Bouowmx]

NVIDIA GeForce GTX 1080 Ti "power save" mode:

Off-topic and flamebait are not allowed here
Markfw
Anandtech Moderator

 

[darkswordsman17]

The only reason I checked Vega was to scalp at this point. $499 was the only MSI card. Me and @Phynaz lol'd at the "token" MSRP card there as a place holdler for the launch.

There's just no point in complaining about low quantities because we don't actually know just how low the stock is.

Finally... a quick economics lesson for all those who are confused as to "Who is buying Vega".
This is actually not hard to explain using economic theory.
In the picture below we have the demand curve.
You can see
Vega MSRP
Slightly Inflated MSRP
Price Vega would sell at if Markets functioned efficiently

At MSRP, Vega would sell a large quantity. However, that quantity isn't available. The inflated price, they'd still sell a decent quantity. But it's STILL not there. The actual quantity is lower. For the quantity supplied, it turns out the amount of buyers is few, but they'd all pay FAR higher prices because of their position on the demand curve. So actually they'd be willing to pay $1000+ (or whatever) and are getting Vega at what they consider to be a steal.

This has nothing to do with your opinion on Vega, it's just basic economics. Anyone is welcome to PM me for a further explanation.
If you'd like to learn on your own:
Economics in One Lesson: The Shortest and Surest Way to Understand Basic Economics
https://www.amazon.com/Economics-One-Lesson-Shortest-Understand/dp/0517548232

Cut the BS condescension (yet another reason why this subforum straight up sucks, how regularly we get "lessons" from random a-holes; were you all out of car analogies or something because those are always fun...). There are plenty of people here that do much more in depth economic analysis for a living so you copping this type of insolent tone is, well sadly all too typical for this subforum, but definitely should be taken as outright insulting. I doubt there's just about anyone here that hasn't had a basic econ course even.

If its as you claim, then all the Vega FE cards are out of stock too then? And all the cards being listed for over $1000 are selling out there as well?

Or could you just be rampantly speculating just because? Nah, that never happens on here.

 

[tential]

Cut the BS condescension (yet another reason why this subforum straight up sucks, how regularly we get "lessons" from random a-holes; were you all out of car analogies or something because those are always fun...). There are plenty of people here that do much more in depth economic analysis for a living so you copping this type of insolent tone is, well sadly all too typical for this subforum, but definitely should be taken as outright insulting. I doubt there's just about anyone here that hasn't had a basic econ course even.

If its as you claim, then all the Vega FE cards are out of stock too then? And all the cards being listed for over $1000 are selling out there as well?

Or could you just be rampantly speculating just because? Nah, that never happens on here.

I never stated a hardprice.
Fe is another product.
Equilibrium price could be $850 thus making everyone who bought it happy.
It was purely just the theory in the most basic terms. I'll pm you further to explain since it seems that some of this went over your head for you to be fixated on the $1000 price point.

Please be willing to accept that there are many reasons why people are happy buying Vega and that's just a basic economic reasoning behind why. You don't have to value Vega, but many do, and hence why it's sold out. This isn't rampant speculation, just the pure underlying reasoning why it happens. The speculation comes later, to explain this underlying reasoning.


Thanks for your polite remarks, they foster a great environment for discussion here!

Edit:also please be sure to send those names of those who do great economic analysis on here my name. I'd love to hear their thoughts on this in a private setting.

 

[Madpacket]

Intel really isn't the fierce competitor that NVidia is. Relatively speaking, Intel was practically standing still waiting for AMD to get it's act together, NVidia is Not. So short of a miracle, I don't expect AMD to close the gap with NVidia.

AMD is trying to make a Jack of all trades (master of none) Video card, while NVidia makes HPC and Gaming focused cards, winning both markets.

While I agree Nvidia is more tenacious than Intel, no one should count out AMD of either space.

Navi is Raja's baby, Vega is just another iteration of GCN with some hardware bits not even being utilized. Look what they're able to accomplish with limited resources. Ryzen and Threadripper is a miracle compared to what most were expecting from a CPU after Bulldozer, and they did that with what 1/10th the R&D budget of their competition?

Now with AMD actually making money, and having overall better value consumer CPU's than Intel, company morale must be up, bonus's will be paid out, and employees will likely try their best to keep this momentum going.

Vega's a disappointment no doubt, but Nvidia has no intention to release Volta anytime soon (why would they?). AMD has (just) enough to compete with Nvidia's lineup minus the top dog 1080 Ti.

Prices of Vega will inevitably drop and drivers should mature to keep RTG in the mix for some time.

AMD is fighting battles on two fronts (Intel and Nvidia) but this gives them some serious advantages like APU's, console design wins, etc. They even somehow convinced Apple to take on their products.

I'm not worried about AMD staying solvent, it was pretty bleak for a while but they're in a much better position these days.

Sure I'm not that happy with Vega's power consumption, and the price should be lower but I bought one anyway. Geforce 1080 speeds for my 1440P IPS Freesync monitor that I purchased a while ago should keep me happy for a few more years.

The hydro costs are mostly overblown IMHO plus you can limit the cards heat output using features such as FRC and Chill during these warmer months if necessary. Yes that's annoying to have to enable but not the end the world.

Of course if I had paid an extra $300.00 CAD for an equivalent Gsync panel I would have bought a Geforce 1080/Ti instead, but I'm sure I'll survive

 

[tential]

Why is it that Intel wasn't tenacious? I don't think Intel should have offered more cores. It would have hurt their profit margins for no reason.

That's doesn't mean Intel was resting on chip development. They were definitely trying to increase IPC. Do you think Intel pours that money into R&D got fun? This narrative that Intel stalled cpu development is the largest joke ever

What Nvidia is doing is execution on another level. Like I said when the gtx 1070 came out. Nvidia is executing on a level in that I'd rather own their stock than their gpu.

Nvidia increasing performance and efficiency generation after generation in this manner is just beautiful to see..

It's not that Intel, amd, you can throw snapdragon chip makers in there as well, aren't trying. Everyone is. But if you compare that to a benchmark of Nvidia's current performance, that's a Tom Brady level of performance.

If you compare Vega as an iteration over past products in comparison to Intel cpus it's not bad.

Only the future will show us who continues executing and who changes their tune.

 

[ozzy702]

NVIDIA GeForce GTX 1080 Ti "power save" mode:

1080tis are pretty damned efficient under about 1400mhz. I have mine clocked low because I'm already at the power/heat threshold I can stand/use and it's amazing to watch the power consumption drop as you lower the power limit. For that matter, all of my Pascal cards are insanely efficient under about 1400mhz. NVIDIA has done an amazing job year after year of increasing efficiency. As much as I'd love to see AMD narrow the gap I don't see that happening at all until at least 2020.

If AMD goes really really wide and clocked low with Navi maybe they can narrow the gap on efficiency (Nano is the only time we've seen this strategy from AMD) but balancing that with performance/price is going to be a challenge. If I had to bet, I'd bet on the gap widening between NVIDIA and AMD on efficiency because of AMD's inability to execute and NVIDIA's flawless execution since the 500 series.

 

[lobz]

Y

Sent from my VTR-L09 using Tapatalk

 

[Unreal123]

I told you guys long time ago that Vega will be a biggest failure in GPU history. I am still laughing at some users who are trying to justify Vega price ,where at the same price they can get GTX 1080 Ti ,which has 35% more performance, better overclocking, 3 GB more ram, cooler, better power consumption.

One more thing i notice is that the whole RTG is silent on twitter ,where as they post 100s of tweets everyday normally ? Why they dead silent after vega launch?

 

[IntelUser2000]

I told you guys long time ago that Vega will be a biggest failure in GPU history.

May I ask how? My question isn't about doubting whether you really knew. You just act like you knew for sure that it would be this bad.

 

[Unreal123]

May I ask how? My question isn't about doubting whether you really knew. You just act like you knew for sure that it would be this bad.

Because on January AMD demo Doom running on VEGA. Looking at AMD history, they always shows a Demo with the best case possible and it is 100% confirm it will perform below or on par with GTX 1080 after showing that demo.

Another lesson is that AMD slides of GPUs features always opposite of performance. 99% of AMD slides about features are always useless or BS just like fury X, bulldozer. Now VEGA.

 

[IntelUser2000]

Because on January AMD demo Doom running on VEGA. Looking at AMD history, they always shows a Demo with the best case possible and it is 100% confirm it will perform below or on par with GTX 1080 after showing that demo.

Another lesson is that AMD slides of GPUs features always opposite of performance. 99% of AMD slides about features are always useless or BS just like fury X, bulldozer. Now VEGA.

But which company doesn't use that sort of tactic? Heck, what do you do when you go for a Job interview or make a Resume? Put your best foot forward. Yes, I'm pretty sure everyone has sucky parts. What about Ryzen though? That turned out well.

That's doesn't mean Intel was resting on chip development. They were definitely trying to increase IPC. Do you think Intel pours that money into R&D got fun? This narrative that Intel stalled cpu development is the largest joke ever

What Nvidia is doing is execution on another level. Like I said when the gtx 1070 came out. Nvidia is executing on a level in that I'd rather own their stock than their gpu.

Well, I guess they could have given us more cores. At least for HEDT. Coffeelake seems more of a deliberate plan. Even then, if their 14/10nm didn't get delayed, we'd have seen 8 core Cannonlake on 10nm last year. Couple that with Gen 10 iGPU they'd have zero reason to worry about Ryzen and Raven Ridge.

Cores aren't a panacea in CPUs though. So the comparison in % in comparison to GPUs aren't exactly fair. Increased shaders are always better. Increased CPU cores not so.

But, I do agree. Nvidia has better execution. Likely because they put importance on their employees. High employee retention rate and CEO approval rate gets reflected on the products they make.

 

[Trumpstyle]

The only reason I checked Vega was to scalp at this point. $499 was the only MSI card. Me and @Phynaz lol'd at the "token" MSRP card there as a place holdler for the launch.

There's just no point in complaining about low quantities because we don't actually know just how low the stock is.

Finally... a quick economics lesson for all those who are confused as to "Who is buying Vega".
This is actually not hard to explain using economic theory.
In the picture below we have the demand curve.
You can see
Vega MSRP
Slightly Inflated MSRP
Price Vega would sell at if Markets functioned efficiently

At MSRP, Vega would sell a large quantity. However, that quantity isn't available. The inflated price, they'd still sell a decent quantity. But it's STILL not there. The actual quantity is lower. For the quantity supplied, it turns out the amount of buyers is few, but they'd all pay FAR higher prices because of their position on the demand curve. So actually they'd be willing to pay $1000+ (or whatever) and are getting Vega at what they consider to be a steal.

This has nothing to do with your opinion on Vega, it's just basic economics. Anyone is welcome to PM me for a further explanation.
If you'd like to learn on your own:
Economics in One Lesson: The Shortest and Surest Way to Understand Basic Economics
https://www.amazon.com/Economics-One-Lesson-Shortest-Understand/dp/0517548232

I think you're confused. What if retailers need to sell the card at minimun 650$ to even make a minor profit(MSRP is meaningless). Does this not make your whole point pointless