I didn't say that. They will still have to work with their customers to validate the hardware. But that is a flat cost cost really, they don't have any real development costs for "parts that serve a mission critical need". OEM's are basically free to integrate the hardware that they are comfortable onto the platform rather than fit their requirements around AMD's implementation (like they do with Intel). This means quicker validation. So basically they just have AMD's design consultants, on site guys to help develop a validation suite, and whatever the supply of CPU's they need for the testing (which might be the largest portion of costs needed to apply to actual EPYC shipment costs). If they continue to use the platform for several generations, that will only quicken the ability for the OEM's to test what they need.
EPYC will have/had significant validations costs well above and beyond what was required for Ryzen or even TR.
I agree. Which is why I find it so peculiar that Epyc is shipping before TR.
Well both ECC and iommu both weren't "verified"/required by the platform spec but they weren't disabled in the CPU, so you just had to buy a MB that supported both of them, like I did.How so?
Why an 8 core TR when there is the 8 core Ryzens?I personally expect 8/12/16 core TR's. I saw some chart with in-between core counts but I don't see it myself based on the CCX configuration. I have no specific paper or source though.
Why a 6 core SKL-X when a 6 core CFL-S will launch shortly after?Why an 8 core TR when there is the 8 core Ryzens?
Well, if the hints in AT article about those ASRock X399 boards and elsewhere are anything to go by, this 4000 contact LGA socket is going to be expensive. And part of that must be because of volume. So one way to get the socket price down for both TR and Epyc is surely to sell more boards. If an 8C part with at least some extra features increases the volume the price for TR and Epyc boards comes down. However, the extra PCIe lanes and memory channels on a dual die package implies a 2+2 & 2+2 to reach 8C which with the rumoured 80% yields means they will most likely have to cripple perfectly good parts.Why a 6 core SKL-X when a 6 core CFL-S will launch shortly after?
Why a 4 core KBL-X when... oh well, let's avoid this mess.
If the 8 core TR keeps the increased PCIe lanes + memory channels... it could make sense somewhat. If it doesn't then it makes little sense indeed.
However, the extra PCIe lanes and memory channels on a dual die package implies a 2+2 & 2+2 to reach 8C which with the rumoured 80% yields means they will most likely have to cripple perfectly good parts.
It doesn't make much sense to put only 1 Zeppelin core in the large package when you can move back to R7 instead for a much smaller package (and cost) for the 8 core. The cores are essentially the same.Why a 6 core SKL-X when a 6 core CFL-S will launch shortly after?
Why a 4 core KBL-X when... oh well, let's avoid this mess.
If the 8 core TR keeps the increased PCIe lanes + memory channels... it could make sense somewhat. If it doesn't then it makes little sense indeed.
Well that and most people don't think the 10 and 14 cores are possible. AMD is doing the anti Intel and doesn't seem to be targeting any overlapping product lines with the exception of their Pro line. I get peoples desires. wanting a high clocked 8 core for $300-$400 that they can plunk down on an almost same cost mobo and have what seems like unlimited memory bandwidth and lanes. But look at the 12c pricing its just slightly above 2x R5 1600 pricing, the 16 cores just higher than 2x 1700x pricing. Offering an 8 core CPU will just create havoc on the pricing basically best case scenario it's $400, causes pricing confusion between the whole R7 lineup, completely cuts into the margins on the TR offerings that people have already been saying are "to small" or "AMD hates money" and frankly not needed. If the 10core CPU's are real and are hovering around $500+ is a completely butchered 2 die chip really worth it, when you are talking about whats supposed to be a premium platform. An 8 core Threadripper is really not that much better than Intel offering KabyLake X on 299 (well it's a lot better, but still not a smart move). The end point being that I don't think an 8c TR brings in more purchasers the ones that would buy that would probably still buy AMD, they would just get an R7 or TR 10c. Though as a bonus it would get more people on X399 which doesn't really help AMD much, but would increase the likelihood of a followup purchase on TR4 as opposed to upgrading to Intel later. Since the rest of the platform is so expensive people would be more likely to upgrade rather than dump the system.Most reports show a low of 10 cores in TR followed by 12, 14 and finally 16 cores. Nowhere did I find an 8 core TR.
Why an 8 core TR when there is the 8 core Ryzens?
Most reports show a low of 10 cores in TR followed by 12, 14 and finally 16 cores. Nowhere did I find an 8 core TR.
Why an 8 core TR when there is the 8 core Ryzens?
Has there actually been an 8 core TR mentioned anywhere by AMD?
Your response doesn't invalidate my question lolNo, only up to 16 cores.
I agree. Which is why I find it so peculiar that Epyc is shipping before TR.
I would think the sweet spot would depend on cost. If and there is some evidence it doesn't exist a 10c was available for $550, would that be just as good as getting an 8c. Specially since unlike Intel's offerings we know that the extra cores really isn't going to affect clock speed.That could be the sweet spot for quite a few. 8 cores with tons of lanes. That's literally why some didn't buy Ryzen. But not sure if AMD would bring out that sku depending on yields.
Your response doesn't invalidate my question
Is that a heavy Scottish accent? It's pretty brutal to listen too. I gave up.
I've been a long time subscriber to his channel. I really like the content he puts out.ok - so an accent can deter you from getting some very interesting information? I dont care how they sound as long as it goes somewhere
Snowy Owl is supposedly an exact server counterpart to Threadripper - 2 Zeppelin dies.