Threadripper: The inside story

[cj09beira]

4 hours ago, emosun said:

wow. too bad they sat around and did nothing for several years prior to that. while intels "computer enthusiasts" got bored from waiting for someone to actually compete.

i dont really think intel is very concerned , amd has a history of not really staying on top of things , and a history of never really being at the cutting edge of anything. short of making cheap versions of cpus that already exist , amd doesn't do much better than that.

you must be new to the tech world, amd was the first to have: 2 core cpu and 64bit cpu, their first Athlon and the opteron were beasts at the time and were better than the intel counterparts, the problem was that intel has been systematically bribing oems and retailers to not sell amd systems / cpus,

there have been a lot of cort cases because of this but they take too long and in the end amd was forced to sign a deal where they would get some money to keep their mouth shut ( they were really almost out of money).

see here for more info:

 

[ApolloFury]

6 hours ago, mynameisjuan said:

"But a Chevy Cruze has a better price/performance than a Hellcat. It must be better!" -AMD fanboy buying a racing car

If you are going to use car analogy...Back in 2012 A Nissan GTR outperforms the Ferrari 458, Audi R8, Mercedes AMG, Porsche 911, while costing less, have more seats, larger in size and more fuel efficient. 

 

[GDRRiley]

4 hours ago, cj09beira said:

 

 

last time I used this I got called an AMD fan boy.

[RGProductions]

7 hours ago, mynameisjuan said:

"But a Chevy Cruze has a better price/performance than a Hellcat. It must be better!" -AMD fanboy buying a racing car

No that's wrong. That's like saying a G4400 has better price/perf than the $2000 CPUs. This is a lot more along the lines of buying a miata instead of an ecoboost mustang. You arent losing much, and its a lot cheaper.

[Guest]

9 hours ago, emosun said:

wow. too bad they sat around and did nothing for several years prior to that. while intels "computer enthusiasts" got bored from waiting for someone to actually compete.

i dont really think intel is very concerned , amd has a history of not really staying on top of things , and a history of never really being at the cutting edge of anything. short of making cheap versions of cpus that already exist , amd doesn't do much better than that.

Cause making a new design takes 8 hours lol bul!dozer sucked they had to go back and not do things such as share important resources with 2 cores and only have one FP unit for 2 cores you know simple things that one can learn in a community college CPU architecture class. 

 

Hell many predicted bulldozer disaster before release but instead we had people from Amd trying to claim they were wrong. I remember I contacted jf-amd and he said bulldozer IPC would be higher then phenom ii despite me showing how that made no sense when they took away resources for one core and shared important aspects such as l2 cache fetch and decode and increased the size of the pipeline. Well 8150 released and it had less IPC then phenom ii.

Edited September 6, 2017 by Guest

[Jito463]

7 hours ago, samcool55 said:

it's important to hire people that have a passion for their job even if they are below average.

A bit of a backhanded compliment.

7 hours ago, Dylanc1500 said:

18 cores was done because of the different silicon dies (10,18, 28). I don't know what you mean by 12 core, if you could explain that a little better, I apologize.

 

like I had said it was released earlier than planned, they had plans to release but not when they  ended up. I can't really go to in depth with that.

7 hours ago, TOMPPIX said:

got a source on that?

I'd like to know that, too.  There was absolutely no mention of anything above 10 core until the beginning of the year, and after the TR announcement then there was suddenly 14-18 core (without any details, even).

7 hours ago, Dylanc1500 said:

Not legally.

What a cop out.

[Guest]

7 hours ago, Dylanc1500 said:

That's very dependent on the workload, clock speed can have a great effect on multicore workloads. (Example of a workload I'm personally familiar with) If you have a database running transaction records, which can saturate ever core you give it, you increase the clock speed and every core can finish its individual workload quicker and move on to the next task.

They didn't even show these products at first they wearnt even on the chart you know Linus kind of made a popular video of this

[Drak3]

1 hour ago, GDRRiley said:

last time I used this I got called an AMD fan boy.

Stupid AMD fanboy, using facts and logic to back up your fact and logic based opinion.

 

Kappa.

44 minutes ago, RGProductions said:

This is a lot more along the lines of buying a miata instead of an ecoboost mustang.

I'd liken it more to getting a standard Ford F150 instead of the Raptor or Lightning. All are interchangeable until you need explicitly what the Raptor and Lightning are/were good at, but the F150 didn't have the premium to do those specific things.

[mr moose]

Do some of you actually think Intel are capable of puling an extra 4 -8 cores out of their arse in such a short time with no planning or R+D?  Of course they were working on it prior to TR, given the x299 release debacle it's fairly obvious they rushed it out the door, but to assume it didn't exist (in planning or otherwise) prior to TR is a bit of stretch.

 

TL:DR: is anyone really surprised that professionals managed to do their job and design a processor worth selling?  It's really kinda insulting to call engineers "enthusiasts" or to insinuate AMD got TR out of employee luck. 

[Drak3]

Just now, mr moose said:

but to assume it didn't exist (in planning or otherwise) prior to TR is a bit of stretch.

Given Xeon-W's recent announcement, it's safe to assume 18c 2066 CPUs were in the works, but not for X299. The time between X299's launch and the 7980XE's release is enough time for Intel to enable and disable features, and do a fair amount of validation.

[Godlygamer23]

9 hours ago, Dylanc1500 said:

If you want top tier enthusiast performance you have to pay top tier enthusiast price. As far as being under water that could be a lot of different things and isn't currently known. Being under water doesn't automatically mean better than air.

To be fair, "liquid cooling" and "air cooling" both require air and liquid to actually cool the components they're mounted to. Once you start learning about them in-depth, the lines that separate them start to blur together, mainly because they use the same mechanics to work.

[ARikozuM]

8 hours ago, TVwazhere said:

Because companies like Autodesk dont want to reinvest rebuilding their code from the ground up to include multiple cores in the part modeling, assembly, construction (revit), etc, which from what I've been told, is what it would take for programs like Inventor (and solidworks I think) to use multi cores when doing routine functions.

Autodesk does use multiple cores. The issue lies in the fact that most functions are only able to saturate a single core due to the user not being fast enough to make use of those cores since they're done sequentially rather than piece-meal. If you transform a project though, you will see more usage from your cores, but again, once the function is done, it's done.

 

9 hours ago, Dylanc1500 said:

As they say in racing, if you want to play with the big boys, you have pay like the big boys.

You mean I can't use a saucepan as a differential cover?

[ARikozuM]

7 minutes ago, Godlygamer23 said:

To be fair, "liquid cooling" and "air cooling" both require air and liquid to actually cool the components they're mounted to. Once you start learning about them in-depth, the lines that separate them start to blur together, mainly because they use the same mechanics to work.

Why do I feel like your posts have been more scientific since you've adopted Albert Einstein as a user icon?

 

Edit: Where would vapor chamber cooling fall into?

[Godlygamer23]

5 minutes ago, ARikozuM said:

Why do I feel like your posts have been more scientific since you've adopted Albert Einstein as a user icon?

 

Edit: Where would vapor chamber cooling fall into?

Not sure how that's changed since I changed my user icon. Couldn't tell you.

 

Vapor chambers use the same method of cooling as heatpipes according to this website:

http://celsiainc.com/blog-heat-pipes-and-vapor-chambers-whats-the-difference/

 

They're both phase change based cooling, where the liquid is transformed into a gas. So really, "air coolers" as we see them, are two-phase coolers.

[Drak3]

11 minutes ago, Godlygamer23 said:

To be fair, "liquid cooling" and "air cooling" both require air and liquid to actually cool the components they're mounted to. Once you start learning about them in-depth, the lines that separate them start to blur together, mainly because they use the same mechanics to work.

Technically speaking, liquid cooling is ultimately just a higher heat capacity variant of air cooling, as the ultimate end goal is using air to remove heat from the heat sink.

 

Unless you dump the rads in ice water. Then, it is absolutely liquid cooling

5 minutes ago, ARikozuM said:

Where would vapor chamber cooling fall into?

Yes.

[ARikozuM]

1 minute ago, Godlygamer23 said:

Vapor chambers use the same method of cooling as heatpipes according to this website:

http://celsiainc.com/blog-heat-pipes-and-vapor-chambers-whats-the-difference/

They're both phase change based cooling, where the liquid is transformed into a gas. So really, "air coolers" as we see them, are two-phase coolers.

Thanks, I was just reading up on some air coolers and was surprised how many use heatpipes without gas/liquid inside. 

[mr moose]

19 minutes ago, Drak3 said:

Given Xeon-W's recent announcement, it's safe to assume 18c 2066 CPUs were in the works, but not for X299. The time between X299's launch and the 7980XE's release is enough time for Intel to enable and disable features, and do a fair amount of validation.

You can only enable, disable and validate something that already exists.   I think this forum just gets too hung up on basic ideals.  If TR can be designed from existing R+D and made into a usable product then so can x299.  The idea that an entire product lineup is a whim or knee jerk reaction is disingenuous to any companies planning and R+D.   Sure they move around release dates and marketing changes.  Prices also drop in reaction to competition etc, but not whole product development.  Neither AMD nor Intel (or even Nvidia for that matter), can pull a whole new processor/chip out of their arse in a few months when typical design stages are in years. 

 

 

 

[mr moose]

13 minutes ago, ARikozuM said:

You mean I can't use a saucepan as a differential cover?

 

So long as it's made from 4130.  Gotta think about weight.

 

 

[Drak3]

22 minutes ago, mr moose said:

You can only enable, disable and validate something that already exists.   I think this forum just gets too hung up on basic ideals.  If TR can be designed from existing R+D and made into a usable product then so can x299.  The idea that an entire product lineup is a whim or knee jerk reaction is disingenuous to any companies planning and R+D.   Sure they move around release dates and marketing changes.  Prices also drop in reaction to competition etc, but not whole product development.  Neither AMD nor Intel (or even Nvidia for that matter), can pull a whole new processor/chip out of their arse in a few months when typical design stages are in years. 

 

 

 

Read it again, and point out where I said that Intel pulled out a totally new processor out of their asses.

[mr moose]

Just now, Drak3 said:

Read it again, and point out where I said that Intel pulled out a totally new processor out of their asses.

I never accused you of saying that. 

But you did say this in reply to me:

57 minutes ago, Drak3 said:

Given Xeon-W's recent announcement, it's safe to assume 18c 2066 CPUs were in the works, but not for X299. The time between X299's launch and the 7980XE's release is enough time for Intel to enable and disable features, and do a fair amount of validation.

 

Did you mean something else? I am a little confused now.

[Drak3]

Just now, mr moose said:

Did you mean something else? I am a little confused now.

No, simple explanation of planned product being adapted to become an unplanned product.

 

1 minute ago, mr moose said:

never accused you of saying that.

Y'all mofos really should say something like "to build off this," when elaborating a point that is heavily worded as a specific correction to the person you're replying to.

[mr moose]

3 minutes ago, Drak3 said:

No, simple explanation of planned product being adapted to become an unplanned product.

 

 

Yes it could be that, but wasn't x299 on the road map June last year? that's a long time before TR for development.

3 minutes ago, Drak3 said:

Y'all mofos really should say something like "to build off this," when elaborating a point that is heavily worded as a specific correction to the person you're replying to.

Then why didn't you when replying to me in the first place?   mofo indeed...

[Drak3]

Just now, mr moose said:

 

Yes it could be that, but wasn't x299 on the road map June last year? that's a long time before TR for development.

X299, and up to either 10c or 12c, but certainly not 18c. Whereas we just got the announcement of Xeon-W and that is up to 18c and has been as far as we know.

 

1 minute ago, mr moose said:

Then why didn't you when replying to me in the first place?

Probably because my reply wasn't worded in such a way that explicitly looks like I'm actually correcting you. I didn't say anything that directly implies that I'm correcting something you said. Whereas, the first thing you said was:

 

45 minutes ago, mr moose said:

You can only enable, disable and validate something that already exists.

In responce to my statement that Intel can simply enable and disable features from a fully planned product, in order to adapt it into an unplanned product. You went as far as to even bold it when you quoted me.

[mr moose]

2 minutes ago, Drak3 said:

X299, and up to either 10c or 12c, but certainly not 18c. Whereas we just got the announcement of Xeon-W and that is up to 18c and has been as far as we know.

 

Probably because my reply wasn't worded in such a way that explicitly looks like I'm actually correcting you. I didn't say anything that directly implies that I'm correcting something you said. Whereas, the first thing you said was:

 

In responce to my statement that Intel can simply enable and disable features from a fully planned product, in order to adapt it into an unplanned product. You went as far as to even bold it when you quoted me.

and now we go in circles.

[Drak3]

Just now, mr moose said:

and now we go in circles.

Given interactions with some other members in the past, generic circles aren't that bad.