AMD Rome Second Generation EPYC Review: 2x 64-core Benchmarked • Hacker News Board

Re: AMD Rome Second Generation EPYC Review: 2x 64-core Benchmarked

#171

Post by AgloeDreams » Thu, Aug 08, 2019, 2:20 PM UTC

_delirium wrote: post #96

bri3d wrote: post #89

Earlier quoted context omitted.

Zen was originally led by the same person as Opteron/K8/Hammer, Jim Keller... who now works for Intel. I think the situation is a bit different this time around though as AMD’s bet on TSMC is paying off in a major way while Intel continue to flounder in the fab space.

> AMD’s bet on TSMC is paying off in a major way Going back a bit further, AMD spinning off GlobalFoundries and then shopping around for fabs on the open market is definitely looking like a very good decision with hindsight. GF has also, since then, run into problems rolling out a next-gen node, and eventually cancelled their 7nm. Hard to say how much you can credit AMD for foresight there vs getting lucky, but being…

What's fun about this is now Intel's world is on fire from a single source. The only way AMD could have guess that the TSMC process would improve so much is if they guessed that Apple would get into their own chip design (which only became clear in 2011-2012 and showed results in ~2014-2015) and would bankroll TSMC's 10 and 7.

Re: AMD Rome Second Generation EPYC Review: 2x 64-core Benchmarked

#172

Post by ksec » Thu, Aug 08, 2019, 2:24 PM UTC

Nokinside wrote: post #135

DuskStar wrote: post #2

> For those with little time: at the high end with socketed x86 CPUs, AMD offers you up to 50 to 100% higher performance while offering at a 40% lower price. Unless you go for the low end server CPUs, there is no contest: AMD offers much better performance for a much lower price than Intel, with more memory channels and over 2x the number of PCIe lanes. These are also PCIe 4.0 lanes. What if you want to more than 2 T…

Xeon has been making 50–70% profit margin depending on the Xeon type. Intel's profit margins will take a hit when they are forced to cut prices. It's unlikely that AMD has enough manufacturing capacity to fill all demand, so Intel will still make profit. History: AMD had similar upper hand against Intel with their Athlon processor 20 years ago. Intel's transition to the 180 nm process was delayed and AMD's K7 Athlon…

>It's unlikely that AMD has enough manufacturing capacity to fill all demand,

Betting against TSMC, not a very wise move at the moment.

>If Intel force AMD's profits close to zero, it can spoil AMD's technical win.

That is assuming Intel will make a lost in other to hurt AMD, and I am not sure if predatory pricing is legal in US. AMD is not pricing their product to undercut Intel, AMD is pricing their product as it is because their Chiplet strategy allows them to do so while still retaining Industry level margin.

I am pretty sure Intel can afford to lose billions, just like their contra revenue, but this time around the value will be so great I am not sure if their Investor will be happy with it, not to mention their declining margin.

Re: AMD Rome Second Generation EPYC Review: 2x 64-core Benchmarked

#173

Post by paulmd » Thu, Aug 08, 2019, 2:28 PM UTC

thecopy wrote: post #112

paulmd wrote: post #75

Earlier quoted context omitted.

> TSMC's 7nm process is stable, providing excellent yields and has plenty of available capacity Yields in terms of getting lots of operational dies, sure, but one of the underlying problems here is chip quality. They have lots of chiplets but most of those chiplets are not fast enough to hit the clocks advertised for the 3900X. In fact, even a lot of the chiplets sold as 3900X are not fast enough to hit the clocks ad…

>a lot of the chiplets sold as 3900X are not fast enough to hit the clocks advertised for 3900X Do you have some sort of source for this claim?

der8auer has done extensive testing on multiple CPUs (he had 12 samples) and has discussed the topic.

https://youtu.be/WXbCdGENp5I?t=119

A lot of reviewers have noted similar things, but often are working with singular samples and didn't want to make too much of a stink without more data, but the problem is widespread. Out of all of der8auer's CPUs, only one hit its advertised boost clocks, and it was one of the lower-end CPUs with a less ambitious target to hit.

It may be a problem with early AGESA firmware, and silicon quality will definitely go up over time, but at least at this point in time AMD has certainly falsely advertised the clocks these CPUs are capable of achieving.

Re: AMD Rome Second Generation EPYC Review: 2x 64-core Benchmarked

#174

Post by ksec » Thu, Aug 08, 2019, 2:29 PM UTC

devonkim wrote: post #43

chubbyrabbit wrote: post #24

Earlier quoted context omitted.

> I can't buy a Ryzen 3900X without paying substantially inflated ebay prices. I'm experiencing the opposite. There has been so many deals on 3900X that I have to constantly tell myself I don't need an upgrade.

I am genuinely intrigued. I've been going to my local Microcenters in the DC VA area for weeks now and they said they have gotten ZERO shipments in since release day of the 3900X and have a couple 3700X maybe on shelves. This doesn't necessarily completely invalidate my point though - distribution by AMD is clearly needing some work when one region is drowning in 3900X processors and a very wealthy metro area has non…

Or may be ( cough ) someone in your local Microcenters hasn't been ordering to fulfil those stocks.

This strategy is quite widely used in many other industry as well.

Although given it has only been launched for less than a month and demand is actually through the roof ( I have been seen reviews being so Pro AMD, even in the AMD Athon 64 days ). So I think it is simply supply is a little tight while TSMC are working hard.

Re: AMD Rome Second Generation EPYC Review: 2x 64-core Benchmarked

#175

Post by ksec » Thu, Aug 08, 2019, 2:30 PM UTC

Nokinside wrote: post #148

lmilcin wrote: post #143

Earlier quoted context omitted.

Of course, even if they were totally kicking ass it will take many years for AMD to build more capacity to what Intel has. Fabs take astronomic investment and time to build. But, it seems Intel is and has been for quite some time unable to even keep up with competition. It was leader and it seems 10 and 7nm processes completely stopped them in their tracks with regards to competition. I am kind of interested what act…

They made some process technology choices that didn't work out as they planned. The delay of the 10nm process and patching the delay with 14+ and 14++ didn't stop anything else they do in process technology. They have accelerated the path to 7nm (similar to TSMC's 5mm). ps. Intel plans to launch their first discrete GPU in 2020 using 7nm process. That's going to be interesting.

>ps. Intel plans to launch their first discrete GPU in 2020 using 7nm process. That's going to be interesting.

It is 10nm in 2020, defiantly not 7nm. That is scheduled for 2021, if they could make it on time.

Re: AMD Rome Second Generation EPYC Review: 2x 64-core Benchmarked

#176

Post by paulmd » Thu, Aug 08, 2019, 2:32 PM UTC

Hikikomori wrote: post #133

paulmd wrote: post #75

Earlier quoted context omitted.

> TSMC's 7nm process is stable, providing excellent yields and has plenty of available capacity Yields in terms of getting lots of operational dies, sure, but one of the underlying problems here is chip quality. They have lots of chiplets but most of those chiplets are not fast enough to hit the clocks advertised for the 3900X. In fact, even a lot of the chiplets sold as 3900X are not fast enough to hit the clocks ad…

Not reaching advertised boost clocks seem to be related to newer AGESA releases from AMD to motherboard vendors, so I wouldn't blame it on chip quality just yet. These contain bugfixes (RdRand for example) and other changes, but has impacted boost clocks. People running AGESA 1.0.0.2 report reaching boost clocks easily (sustained in single-core tests), while I and others running later releases have issues. New archit…

I've seen plenty of people having problems on older AGESA too. I've seen some people actually have higher performance on newer AGESA. It all depends on your particular sample and setup and how it fits into the boost criteria. Silicon quality still plays a massive role.

So to be clear, older AGESA isn't a magic bullet that is letting all chips "easily hit their rated boost clocks".

It could be cleaned up somewhat in future AGESA releases, and silicon quality will definitely go up over time.

Re: AMD Rome Second Generation EPYC Review: 2x 64-core Benchmarked

#177

Post by ksec » Thu, Aug 08, 2019, 2:34 PM UTC

mktmkr wrote: post #34

aeleos wrote: post #33

> "We designed this part to compete with Ice Lake, expecting to make some headway on single threaded performance. We did not expect to be facing re-warmed Skylake instead. This is going to be one of the highlights of our careers" Looks like AMD expected Intel would actually start to fight back a few years ago when AMD started the Zen and Rome cores, and AMD has been running full steam ahead since then. Meanwhile, in…

In the “what a time to be alive” category, this is essentially a repeat of Opteron.

I am not seeing any similarities to Opteron era, from Intel's current position, AMD's Roadmap and execution, to TSMC's Roadmap.

Re: AMD Rome Second Generation EPYC Review: 2x 64-core Benchmarked

#178

Post by dbancajas » Thu, Aug 08, 2019, 2:36 PM UTC

ksec wrote: post #172

Nokinside wrote: post #135

Earlier quoted context omitted.

Xeon has been making 50–70% profit margin depending on the Xeon type. Intel's profit margins will take a hit when they are forced to cut prices. It's unlikely that AMD has enough manufacturing capacity to fill all demand, so Intel will still make profit. History: AMD had similar upper hand against Intel with their Athlon processor 20 years ago. Intel's transition to the 180 nm process was delayed and AMD's K7 Athlon…

>It's unlikely that AMD has enough manufacturing capacity to fill all demand, Betting against TSMC, not a very wise move at the moment. >If Intel force AMD's profits close to zero, it can spoil AMD's technical win. That is assuming Intel will make a lost in other to hurt AMD, and I am not sure if predatory pricing is legal in US. AMD is not pricing their product to undercut Intel, AMD is pricing their product as it i…

They can only lower their profit margin if AMD has the same performance as Intel. Seeing AMD has higher performance there is no way customers will buy a cheaper offering with less performance. It's like Intel vs AMD but reversed. For years AMD's offering lower performance lower priced chips but can't break through the market (CPU).

Re: AMD Rome Second Generation EPYC Review: 2x 64-core Benchmarked

#179

Post by gwicks56 » Thu, Aug 08, 2019, 2:43 PM UTC

tempguy9999 wrote: post #131

pdpi wrote: post #124

Earlier quoted context omitted.

They might not be definitive sources, but they can definitely be the canary that triggers an investigation by somebody more authoritative. As it stands, this mine has some dead canaries.

claim: > ...but one of the underlying problems here is chip quality supposed justification: > ...every forum pretty much That's hearsay not actual justification. And intel has a reputation for dirty play.

The justification was for the statement that chips aren't hitting their clocks. That is not a controversial statement.

Re: AMD Rome Second Generation EPYC Review: 2x 64-core Benchmarked

#180

Post by sundvor » Thu, Aug 08, 2019, 2:44 PM UTC

wtallis wrote: post #170

lmilcin wrote: post #156

Earlier quoted context omitted.

I didn't know that. Still, somebody would have to build capacity to handle demand and it won't be much faster just because it is TSMC/GF.

The capacity already exists, if AMD can outbid the guys trying to use TSMC to produce GPUs and smartphone SoCs.

In thrilled to see all the good news coming from AMD; that sounds really good for the 3950X I'm eyeing later this year.

I've had enough of the issues with my 6850k.