Will AMD release any high performance GPU's in Q2 or Q3?

[Zenny]

I'm guessing somewhere between 1080 and 1080Ti numbers for a 1080 price. Very nice, but way too late.

[Gaidax]

I seriously doubt this will be price competitive with all that HBM and even if it is, Nvidia will for sure launch a price war which AMD can't hope to win, simply because Vega is such an expensive solution, while Nvidia seem to be churning out Pascals like no tomorrow for a year now, so for sure they are cost efficient for them now.

[mrgreenthump]

I seriously doubt this will be price competitive with all that HBM and even if it is, Nvidia will for sure launch a price war which AMD can't hope to win, simply because Vega is such an expensive solution, while Nvidia seem to be churning out Pascals like no tomorrow for a year now, so for sure they are cost efficient for them now.

290x costed $144 in parts, Fury X costed $187, so overall it's not that much of an issue. Sure AMDs margins will be lower, but when both companies are selling their cards at 2-3 times profit over manufacturing costs, it's hardly going to kill either.

In short, I just cant see Nvidia selling the1080ti at $200.

[Gaidax]

290x costed $144 in parts, Fury X costed $187, so overall it's not that much of an issue. Sure AMDs margins will be lower, but when both companies are selling their cards at 2-3 times profit over manufacturing costs, it's hardly going to kill either.

In short, I just cant see Nvidia selling the1080ti at $200.

Components is only part of the issue. Assembly, defects and development is a huge chunk of why cards cost so much. It's not like they took random shit off the shelf, glued it together and called 1080Ti/Vega.

Nvidia won't be selling 1080Ti for 200 bucks, but they can sell it for $549 no problem if needed and I am certain that full consumer Vega trying to match that will not be feasible for AMD that simply can't stop bleeding money every quarter, while Nvidia swims in it.

Another thing I am certain about, Nvidia ain't going to be giving up market share, so it's a case of 1+1 - if Vega will end up troublesome, they will just choke it by adding crazy value to their stuff.

[mrgreenthump]

Components is only part of the issue. Assembly, defects and development is a huge chunk of why cards cost so much. It's not like they took random shit off the shelf, glued it together and called 1080Ti/Vega.

Nvidia won't be selling 1080Ti for 200 bucks, but they can sell it for $549 no problem if needed and I am certain that full consumer Vega trying to match that will not be feasible for AMD that simply can't stop bleeding money every quarter, while Nvidia swims in it.

My point was more towards they are not going to lower their price to a point where AMD can't make any profit. HBM2 is more expensive than GDDR5X yes, but it is only a part of what goes into the production of a full card etc.

[Snuke]

Has AMD released any high performance GPUs that competed at the top of the line level in the last decade? Honest question...

[Life-Binder]

I think top full Vega will be between 1080 & 1080Ti in perf with a price of $600 (+/- $50)


I dont see them being able to sell cheaper then that, they aint a charity (big die + HBM2 costs etc. .. not less then $550), but at the same time I dont see it flat out matching 1080Ti in OC vs OC, so they wont be able to put anything close to $700 price tag on it

[Butler to Baby Sloths]

Has AMD released any high performance GPUs that competed at the top of the line level in the last decade? Honest question...

Of course they have.

Remember the 4870, 5870 and 7970?

[Shakadam]

Of course they have.

Remember the 4870, 5870 and 7970?

Ahhh those were the days. I bought 2x Radeon 4850's when they were released for about 360€ total (which is what a single mid-level card costs these days.......), that crossfire setup outperformed any single card Nvidia could offer at the time (the top model being the GTX 280), and it was also significantly cheaper than a single GTX 280.

That whole 4xxx series was such a leap for AMD when they were released, one of the best series of GPU's they've ever made imo.

[mrgreenthump]

That whole 4xxx series was such a leap for AMD when they were released, one of the best series of GPU's they've ever made imo.

Weren't that like the last generation before Raja left for Apple? And now that he is back, he NAVI is apparently his brainchild. So that kinda excites me . Wouldn't expect us to go back to high level GPUs costing $200-$300 though, simply because we've shown the willingness to pay $600++ for them. And the same sadly goes for gaming monitors.

[Butler to Baby Sloths]

Weren't that like the last generation before Raja left for Apple? And now that he is back, he NAVI is apparently his brainchild. So that kinda excites me . Wouldn't expect us to go back to high level GPUs costing $200-$300 though, simply because we've shown the willingness to pay $600++ for them. And the same sadly goes for gaming monitors.

VEGA is Raja Koduri's first architecture at AMD after returning.

[Life-Binder]

well

guess we will need to settle in for 1070/1080 competitors, not 1080Ti

http://wccftech.com/amd-vega-10-3dma...hz-core-clock/

http://www.3dmark.com/fs/12296284

this also falls in line with the recent leaked Time Spy score of ~5500-5700+ (thats also Fury X - 1070 level)

AMD Vega 10 3DMark Fire Strike Benchmark Entry Spotted – 687F:C1 Device With 8GB of 700MHz vRAM & 1200MHz Core Clock


AMD’s highly anticipated next generation Vega 10 GPU has been spotted in the 3DMark Fire Strike database. This is the very same GPU that AMD has showcased in numerous performance demos over the past several months.

So, let’s jump straight into the details. First things first, the device ID of the graphics card in the discovered entry is 687F:C1. The device ID matches that of the earliest Vega 10 prototype that we know of. The tested graphics card has 8GB of video RAM running at 700MHz and the card’s GPU clock is a modest 1200MHz.


The Vega 10 687F:C1 Engineering Sample 3DMark Fire Strike Performance Result

The 3DMark Fire Strike entry shows a graphics score of 17801, which is roughly 1400 points more than an R9 Fury X, nearly exactly the same as a Maxwell GTX Titan X and less than a couple hundred points below a GTX 1070.


How AMD Expects Production Ready Radeon RX Vega Graphics Cards To Perform

Remember, the MI25 accelerator is a passively cooled device. Which in reality means 1563MHz is potentially a conservative clock speed for Vega 10 and actively cooled graphics cards should be able to run at higher clock speeds.

Now keep in mind that all of this is based on information AMD had made public early this year based on its expectations. Vega 10 may eventually punch above its weight or fall below those projections, we simply cannot know for certain just yet. What we can safely assume is that Vega 10 will not run at clock speeds below Polaris, based purely on the higher clock speed advantage of the Vega architecture and the maturity level of the FinFET manufacturing process.

Realistically speaking, if AMD manages to successfully hit its 1.56GHz projection, production ready Vega 10 will run at a clock speed that’s 30% higher than what we’re seeing from the 687F:C1 prototype. Performance doesn’t necessarily scale in perfect linearity with the core clock speed however, as other factors play a major role like memory bandwidth and internal architectural bottlenecks that we know very little about at this point.

According to the JEDEC spec HBM2 should scale all the way to 1000MHz/2Gbps but currently available HBM2 stacks are limited to 800MHz/1.6Gbps. With that in mind, a 30% higher GPU clock speed and a 100MHz boost to the memory would move Vega 10 from its current 1070’ish spot to a more competitively favorable position with the GTX 1080.

Radeon RX Vega ( Vega 10 ) Is Set To Be A GTX 1080 Contender

Perfect frequency/performance scaling would put it 7% ahead of the 1080 in 3DMark Fire Strike. A more realistic estimate would put it right around the performance level of the GTX 1080. That is assuming the Radeon RX Vega flagship ends up at around the ~1.56GHz clock speed of the passively cooled MI25 accelerator. So without doubt there’s still more wiggle room for the Radeon Technologies Group if they’re able to push the frequency higher than that. Things like liquid cooling which they’re potentially going for again can definitely help.

A cut down Vega 10 will compete with the GTX 1070. The gap between the GTX 1080 and GTX 1070 is sizable, definitely larger than #80 and #70 cards of past generations. This should help a lower binned Vega 10 — let’s call it Vega 10 Pro — put on a good showing against the GTX 1070.

Before we close here it’s very important to understand that what AMD is attempting with the Vega architecture as a whole does not begin with nor does it end with enthusiast gaming Radeon RX Vega graphics cards.

The new architecture means a lot more than that to the company. Vega’s high bandwidth cache technology is a truly unique memory subsystem. A technology that is poised to shine brightest in professional compute arena, especially in applications that demand massive sets of data. Think AI, machine learning and many forms of data analysis. While the implications of this forward-looking architecture in gaming may be far into the future its potential can be immediately realized in the GPGPU compute space.

With the Vega launch approaching, it’s critical that enthusiasts looking forward to the new GPU hit the pause button for a moment and manage their expectations. If you were expecting a monster that’s going to dethrone NVIDIA’s GTX Titan Xp think again. If you were looking for GTX 1070 and GTX 1080 competition from the Red team to shake things up, later this quarter you’re going to get it.
Vega’s is officially slated for release some time between now and before the end of June according to AMD’s CEO.

[Remilia]

Feels like deja vu, where clock speed of Polaris was supposed to be 800MHz according to leaks.

[Life-Binder]

they specifically address the early ES clock of 1200, but also calc that with current leaked FS/TimeSpy results @ Fury X-1070 levels, with a final retail clock of 1500-1550, the perf will be ~1080 level

seems legit .. certainly doesnt seem like a further additional 30-35% perf (on top of a theoretical 1500+ Mhz result) will spring out of somewhere to match or beat the 1080Ti


might be they plan to later fight the 1080Ti with a dual GPU Vega card, but that comes with all the CF issues, so eh



as for Polaris 800 Mhz - only one leak/rumor said that (not even for the final verison, it was about ES as well) and it was pretty obv fake, noone bought it

[Miyagie]

Iam also tired now of waiting and the MSI Armor GTX 1080 cost atm 460€ so byebye AMD and i hope my 4th Nvidea card wont die after 1 year like the other 3.

Has AMD released any high performance GPUs that competed at the top of the line level in the last decade? Honest question...

295X was very strong.

[Life-Binder]

I honestly dont even count dual GPU cards from either vendor as far as "performance crown" goes

to me they dont exist and only the single GPU ones are the real flagships that compete


single GPU gives stable performance, CF/SLI scaling jumps 0-90% from title to title .. sometimes it can even be negative lol

plus you can always achieve the same with traditional SLI/CF if you want, all it does is save you a PCI-E slot

[Remilia]

they specifically address the early ES clock of 1200, but also calc that with current leaked FS/TimeSpy results @ Fury X-1070 levels, with a final retail clock of 1500-1550, the perf will be ~1080 level

seems legit .. certainly doesnt seem like a further additional 30-35% perf (on top of a theoretical 1500+ Mhz result) will spring out of somewhere to match or beat the 1080Ti

might be they plan to later fight the 1080Ti with a dual GPU Vega card, but that comes with all the CF issues, so eh

as for Polaris 800 Mhz - only one leak/rumor said that (not even for the final verison, it was about ES as well) and it was pretty obv fake, noone bought it

You also need to look at memory. 700MHz for 2xHBM2 stacks is slower than a 4xHBM1 stacks at 500MHz which at that point is definitely memory starved if we're doing a Fiji comparison. Of course don't know how Vega scales, but Polaris for example scales extremely well with memory overclock. Fiji scaled in games for HBM overclock either really well or miniscule depending entirely on the game.

Just wait for the stuff to come out then judge it. =\

Btw this is why rumors are fun. You can find anything you're looking for cause rumors.
https://www.techpowerup.com/233004/a...-than-gtx-1070

[MrPaladinGuy]

well

guess we will need to settle in for 1070/1080 competitors, not 1080Ti

http://wccftech.com/amd-vega-10-3dma...hz-core-clock/

http://www.3dmark.com/fs/12296284

this also falls in line with the recent leaked Time Spy score of ~5500-5700+ (thats also Fury X - 1070 level)

It's WCCTech, no one takes them seriously.

I don't see how we'd get 1080 Ti performance from it unless they delayed it.

Just basing this on history, but according to the specs all that makes the 1080 Ti a Ti is a memory OC, a pretty significant one.

[Remilia]

It's WCCTech, no one takes them seriously.

I don't see how we'd get 1080 Ti performance from it unless they delayed it.

Just basing this on history, but according to the specs all that makes the 1080 Ti a Ti is a memory OC, a pretty significant one.

1080 Ti is a different chip than 1080. Naming is a bit annoying really since 1080 and 1080 Ti aren't related whatsoever. GP102 is what's present in Titan XP and 1080 Ti where 1080 and 1070 is GP104.

[Life-Binder]

It's WCCTech

wccf didnt make the (alleged) Vega Firestrike & TimeSpy results, they only reported it