Technical Issues Affecting Custom AMD Vega Card Production
AMD’s Vega is the first high-end GPU to come out of Team Red in two years, so you would think that its board partners would be quick jump in with new custom products. However, that doesn’t seem to be the case. We’re hearing from sources that AMD’s AIB partners have found several issues with Vega chips that prevent them from creating custom cards for now.
As is often AMD’s strategy, the company rolled out Vega in a staggered release. It first launched the professional workstation-level and air-cooled Vega Frontier Edition in early July, followed by the liquid-cooled version a couple of weeks later. In August, AMD and its partners released the Radeon RX Vega 64 gaming cards, followed shortly after that by the RX Vega 56 cards. All of them were reference designs, and AMD promised custom third-party implementations of Vega later in Q3 ‘17 or early in Q4.
Indeed, most of AMD's partners are working on custom Vega cards--but not all--and for those who are, there seem to be issues.
XFX and Sapphire confirmed that they both have custom boards in the works, but they could not say when they might be ready. PowerColor said that it will have its own custom cards, with mass production scheduled for the beginning of November, but it hasn't yet received the DRAM it needs. (VisionTek didn’t immediately reply to our queries about their future offerings.)
AMD also has partnerships with Asus, Gigabyte, and MSI to build Radeon graphics cards, but these three companies don’t have exclusive deals with AMD. As such, they aren’t driven by necessity and have the luxury of choosing which components to support. We spoke with all three companies, and their responses indicated that their support for the Vega architecture is less definitive than AMD’s exclusive partners.
We already knew that Asus is on board with Vega. The company announced in August that it would be releasing a pair of ROG Strix Vega cards with Asus’ custom cooling solution. Asus confirmed that those cards are still coming, although the release date has been pushed back from September to early October.
Although a Gigabyte rep said it’s likely that the company would be producing a custom Vega card, they would not or could not confirm with 100% certainty that it will. If it does, we likely won’t see it until the end of the year, or later.
Get Tom's Hardware's best news and in-depth reviews, straight to your inbox.
MSI’s response surprised us. The company traditionally offers re-engineered graphics cards with custom PCB designs for all high-end GPU platforms, but it appears to be skipping the Vega lineup. A company representative told us that MSI “won’t be making a custom card anytime soon,” but could offer no additional information.
So what gives? Sources tell us that there is too much variance in the quality of the chips AMD is providing. AIB partners are unable to figure out a stable overclocked GPU frequency that works for all cards, and therefore cannot provide any sort of warranty on factory-tuned cards. Further, there continues to be discrepancies between the temperatures the GPU is reporting and what AIB partners are finding in actual measurements. This is true of the actual GPU and the capacitors below the GPU. We have some follow-up testing that will reveal more about these issues.
Finally, as we reported last month, there have been issues due to the different packages for Vega, making it difficult to efficiently mass produce custom Vega cards. We were seeing Vega with molded and unmolded packages, which we noted impacted package height. We were even seeing a third package--we assume, using SK hynix HBM. As we wrote then:
AIB partners face new challenges, since the HBM2 is about 40 μm lower in the unmolded packages, and the third variant's underfill obviously differs somewhat.For one, the mass production and use of a common cooler for several models must take the applied heat conducting material into account. The thickness must be optimized for the unmolded packages, the viscosity must be high enough, and the resulting contact pressure can't damage anything after being bolted together.
Generally speaking, AIB partners seem optimistic about shipping Vega cards in 2017, and some pointed out that custom Polaris cards came a couple months after the reference card launch. By that timing, we should be seeing some custom Vega cards at the end of September, or at least in October. We’re not getting a strong feeling that will be the case, however.
We’ve reached out to AMD for comment, but the company didn’t immediately reply.
Kevin Carbotte is a contributing writer for Tom's Hardware who primarily covers VR and AR hardware. He has been writing for us for more than four years.
-
JackNaylorPE This is bad .... in not so distant times, nVidia took the top 2 tiers and AMD had the rest ... then it was top 3 ... this round it's top 4. Now AIB partners, most notably MSI who typically had the best AMD offerings ... is walking away from Vega. AMD needs to fix this right away.Reply -
JackNaylorPE Reply20197367 said:hearing from sources what sources?
They quoted them in the article:
XFX and Sapphire confirmed that they both have custom boards in the works, but they could not say when they might be ready. PowerColor said that it will have its own custom cards, with mass production scheduled for the beginning of November, but it hasn't yet received the DRAM it needs. (VisionTek didn’t immediately reply to our queries about their future offerings.)
AMD also has partnerships with Asus, Gigabyte, and MSI to build Radeon graphics cards, but these three companies don’t have exclusive deals with AMD. As such, they aren’t driven by necessity and have the luxury of choosing which components to support. We spoke with all three companies, and their responses indicated that their support for the Vega architecture is less definitive than AMD’s exclusive partners.
Asus confirmed that those cards are still coming, although the release date has been pushed back from September to early October.
Although a Gigabyte rep said it’s likely that the company would be producing a custom Vega card, they would not or could not confirm with 100% certainty that it will.
MSI’s response surprised us. The company traditionally offers re-engineered graphics cards with custom PCB designs for all high-end GPU platforms, but it appears to be skipping the Vega lineup. A company representative told us that MSI “won’t be making a custom card anytime soon,” but could offer no additional information.
Sources tell us that there is too much variance in the quality of the chips AMD is providing.
Obviously, that last statement was crafted to protect their sources as any journalist would. On top of the HBM problem, this makes it very expensive for board partners to make money.
-
Casecutter Ah, The second tier AIB throwing shade like this article. AMD is fulfilling the contract for Apple iMac as that is releasing end of this year. Just like why the Tonga XT never showed, and 380X was not brought in until late. AMD is working to supply the corporate custom clients first, it provides for a known ROI and year end bottom line.Reply -
cryoburner The difference in package design between the different factories doesn't seem like that much of an issue. All AMD has to do is make an agreement with a given card manufacturer that they will receive one design or another. Then they don't have to worry much about making their coolers work well with both formats.Reply -
darthbonehead8 Gigabyte and MSI have realised they can't just re purpose the Pascal cooler which is why Sapphire, XFX and Powercolor (AMD only AIBs) are all onboard, Asus is lifting their game after the fiasco with their Hawaii cards (They reused the Kepler coolers and it sucked).Reply
I mean there's nothing stopping them from bringing out a model with the same clock speeds as the reference card is there? -
bit_user Reply
Then why bother with all the added costs? If it runs at the same clocks, then they won't be able to sell it for (much) more, so it's not worth the effort. Doubly-so, if Vega poses added challenges and issues.20197904 said:I mean there's nothing stopping them from bringing out a model with the same clock speeds as the reference card is there?
I just wonder whether the production lines making the reference boards have the capacity to scale up, or if this could become a bottleneck for getting Vega cards to market in time for the holidays. -
darthbonehead8 Reply20198007 said:
Then why bother with all the added costs? If it runs at the same clocks, then they won't be able to sell it for (much) more, so it's not worth the effort. Doubly-so, if Vega poses added challenges and issues.20197904 said:I mean there's nothing stopping them from bringing out a model with the same clock speeds as the reference card is there?
I just wonder whether the production lines making the reference boards have the capacity to scale up, or if this could become a bottleneck for getting Vega cards to market in time for the holidays.
The majority of AMD AIB cards are at reference clocks anyway, considering AMD has stated RX Vega runs at 1247/945 base so anything over that is considered an overclock right?
I'm sure the marketing teams can whirl up some "30% Overclock" figure to slap on the box while the card is at the same clock speeds as reference.
Most people just want a different cooler for Vega, overclocking your GPU doesn't matter that much outside of benchmark, especially not the little 20-50Mhz bumps that board partners give them. -
hapkiman Ryzen was a long overdue feather in AMDs hat. But Vega.....a big big disappointment and as it seems poorly designed as well. Hopefully they get it done better with their Navi line in 2 years.Reply -
gc9 What does this sentence mean?Reply
"PowerColor said that it will have its own custom cards, ... but it hasn't yet received the DRAM it needs."
I thought AMD's product included VEGA and the HBM memory as a unit preassembled on the interposer.
If Powercolor is waiting on DRAM...
Does that mean that they are waiting on HBM chips, and the board manufacturers are so dissatisfied with the quality of the interposer/HBM/VEGA assembly that they are going to try to assemble it themselves?
Or does it mean that they are waiting on interposer assemblies, and they heard the assemblies are delayed because a shortage of HBM DRAM?
Or does it mean that they are waiting on "better" interposer assemblies, because the memory installed on the interposer assemblies they've received isn't working well enough in their product designs?