Intel Skylake / Kaby Lake

[Zucker2k]

You don't have to tell people what workloads people buying these CPUs ought to run.

Ladies and gentlemen, we just did a full circle!

Not a dig at you Tamz_msc.

 

[tamz_msc]

Ladies and gentlemen, we just did a full circle!

Not a dig at you Tamz_msc.

What is the context of your post?

 

[Markfw]

Intel essentially decimates the Render/Encoding benchmarks which is exactly the type of workload you by extreme core count CPUs for.
https://www.anandtech.com/show/11839/intel-core-i9-7980xe-and-core-i9-7960x-review/10

Just taking the first one, a 20% lead is a fair amount, but decimates is not the word I would use. Now twice the price for 20% performance increase ? If you have the money, sure, but its a poor choice IMO.

And that 20% is only in some things. 1950X actually wins a few, and comes close in quite a few.

 

[SlowSpyder]

Just read Anandtech's review of the 7980XE... nice CPU, very fast, but doesn't pull away far enough or consistently enough to justify its price tag in my opinion.

 

[Dayman1225]

https://videocardz.com/72940/intel-z370-motherboards-unboxing-videos-are-here


Z370 boards unboxed

 

[moonbogg]

Yeah the big Intel chips are fast and awesome, but those prices really tarnish them a lot. Also, Threadripper came first, so Intel is second to the party and missed all the excitement. Fast chips and good to use if you want the best and are willing to pay a hilarious premium for it.
I admit to being shocked that they work on existing boards. I have to admit to being as wrong as a fool can be in thinking Intel would release another board for the big chips. Who else got that wrong? Come on! Man up and admit it! Feels good!

 

[PeterScott]

Just taking the first one, a 20% lead is a fair amount, but decimates is not the word I would use. Now twice the price for 20% performance increase ? If you have the money, sure, but its a poor choice IMO.

As pointed out before, you really can't expect price/performance linearity at this level.

Generally you wouldn't be buying these chips unless you are making money on them. If you are making money on them, a 20% performance delta will quickly pay for itself in a couple of weeks from increased productivity.

Which Rendering/Encoding tests did TR win? There is a difference between winning some obscure benchmark, and winning on the workloads where you are making money.

 

[moonbogg]

Making money with these chips? Exactly what are these "money workloads" you speak of?

 

[Jan Olšan]

7980xe draws less power than 1950x when both are at stock. When both are overclocked, 1950x to 4.0GHz and 7980xe to 4.5GHz, it still draws less power...

What the hell did the guy do with the Threadripper, 1.5V? Sounds kinda bogus (read = not representative) to me.
Look at what PC Perspective measured (edit: link):

The i9-7980XE is at 4,3 GHz, 1.2 V.

Okay, guys, this is a real number. And it’s not even the highest number I measured in our testing, but it is the most apples to apples. When setting voltage to AUTO in the ASUS BIOS and running at 43x multiplier, I measured a power draw of 610 watts with Blender. After backing it down to 1.2v fixed and using Cinebench as our standard power draw test, we get 553 watts. That is 279 watts HIGHER than the 7980XE running at stock settings. Even if you assume that the 7980XE was running at its specified 180 watts TDP in those conditions (which is debatable), that puts this CPU at 459 watts of total power draw.

Even though Threadripper also shoots up when OCing on raised voltage, Skylake-X seems to have much bigger "potential" in this. And note that the CB R15 test doesn't even use AVX(2) or even AVX512

 

[tamz_msc]

Generally you wouldn't be buying these chips unless you are making money on them. If you are making money on them, a 20% performance delta will quickly pay for itself in a couple of weeks from increased productivity.

Again this crap. Sure the only people buying these CPUs are those who only run Ansys Fluent or pay Oracle licensing fees.

Apparently independent software devs and researchers using open-source stuff exist in an alternate universe.

 

[LTC8K6]

Again this crap. Sure the only people buying these CPUs are those who only run Ansys Fluent or pay Oracle licensing fees.

Apparently independent software devs and researchers using open-source stuff exist in an alternate universe.

If they wouldn't make it back, then why would they pay for it?

It seems like a fairly simple calculation.

If it's a hobby, then it's a hobby. Hobbies are generally expensive and there's no expectation of monetary return.

If it's business, then it's business, and it either pays for itself, or it doesn't.

 

[PeterScott]

Making money with these chips? Exactly what are these "money workloads" you speak of?

Commercial use for Rendering/Encoding.

 

[FIVR]

What the hell did the guy do with the Threadripper, 1.5V? Sounds kinda bogus (read = not representative) to me.
Look at what PC Perspective measured:

The i9-7980XE is at 4,3 GHz, 1.2 V.



Even though Threadripper also shoots up when OCing on raised voltage, Skylake-X seems to have much bigger "potential" in this. And note that the CB R15 test doesn't even use AVX(2) or even AVX512

He gets all his technical information from youtube videos, so obviously there will be some mistakes here and there. Don't hold it against him, some people just aren't big readers.

 

[AtenRa]

Even though Threadripper also shoots up when OCing on raised voltage, Skylake-X seems to have much bigger "potential" in this. And note that the CB R15 test doesn't even use AVX(2) or even AVX512

Ive heard close to double that at 4.8GHz on AVX512 loads.

 

[tamz_msc]

If they wouldn't make it back, then why would they pay for it?

It seems like a fairly simple calculation.

If it's a hobby, then it's a hobby. Hobbies are generally expensive and there's no expectation of monetary return.

If it's business, then it's business, and it either pays for itself, or it doesn't.

Try to assign a monetary value to scientific research. Software devs aren't usually paid to do a given amount of work per day, they're paid on the value of the end result.

The only place where this argument works is if you're actually using licensed software. Stop making this argument as if that's the only thing that matters.

 

[Zucker2k]

Making money with these chips? Exactly what are these "money workloads" you speak of?

What do you use Mac and IMac Pros for? Content creation mayhaps?

 

[Zucker2k]

Ive heard close to double that at 4.8GHz on AVX512 loads.

AVX 512 at 4.8Ghz? Where did you pull that one from? Oh, you forgot they achieved that on ln2.

 

[LTC8K6]

Try to assign a monetary value to scientific research. Software devs aren't usually paid to do a given amount of work per day, they're paid on the value of the end result.

The only place where this argument works is if you're actually using licensed software. Stop making this argument as if that's the only thing that matters.

So, they have to make their own determinations. Just like I said.

Is someone forcing Core X on them?

 

[stockolicious]

I don't think Intel really wants to sell the 18 core variant - The pricing is way off and its performance is not always above their own 16 core chip. Intel's 18 core chip must cost a bunch more than Threadripper's 2 x 8 core die's. I think Intel just wanted to make sure the market knew they have the fastest chip (which appears to be true) but I don't think its a Threadripper killer. IMHO - also, If AMD has a real advantage in the CPU market, its as we approach the high end where the cost issue grows for Intel due to their Monolithic approach.

 

[eddman]

He gets all his technical information from youtube videos, so obviously there will be some mistakes here and there. Don't hold it against him, some people just aren't big readers.

Why some members like to insult be disrespectful towards others for no reason. It's from overclock3d.

Doesn't say under what program it was measured but the footnote says CPU+GPU. A game? That can't be it, can it?

 

[FIVR]

Why some members like to insult others for no reason. It's from overclock3d.

Doesn't say under what program it was measured but the footnote says CPU+GPU. A game? That can't be it, can it?

Not an insult (unless saying he is 'wrong' is an insult to you). In fact, I told the guy not to hold it against him for posting fake graphs he got from youtube.

If you'd like to complain about somebody "insulting" others, maybe you should check out his last post...

 

[jpiniero]

I don't think Intel really wants to sell the 18 core variant - The pricing is way off and its performance is not always above their own 16 core chip.

If anything, the other HCC models aren't going to sell. Don't forget about the e-penis factor.

 

[MarkPost]

What the hell did the guy do with the Threadripper, 1.5V? Sounds kinda bogus (read = not representative) to me.
Look at what PC Perspective measured (edit: link):

The i9-7980XE is at 4,3 GHz, 1.2 V.



Even though Threadripper also shoots up when OCing on raised voltage, Skylake-X seems to have much bigger "potential" in this. And note that the CB R15 test doesn't even use AVX(2) or even AVX512

1.5v?

LOL

what a joke of reviewer really

 

[tamz_msc]

So, they have to make their own determinations. Just like I said.

Is someone forcing Core X on them?

Once the initial purchase has been made, how well a CPU like this pays for itself has almost no bearing on the actual performance offered, except in very specific instances, and those instances aren't representative of the kind of users buying these CPUs off retail and etail.

If you want the best performance in something like Ansys Fluent, you'd ask your company to invite tenders for one of these instead.

 

[coercitiv]

It wasn't long ago a Hardware Unboxed review was considered "trash data" because it didn't line up to the silicon world view of some forum members, now Overclock3D - a review site more than 10 years old - is being trashed for publishing power data that doesn't fit right with other folks.

Maybe, just maybe, we should consider sample variance and use data from more reviews to get a more accurate image?! Oh, who am I kidding, Tiny 'Tim' Logan is a YouTube amateur! /s