Intel Haswell-EX an 18-core Leviathan

[IvantheDugtrio]

Per CPU, isn't it? So wouldn't this HELP?

I want to say RHEL licenses are per core though...

Nah the whole enterprise industry refers to CPUs as actual cores (not to be confused with threads). So an IBM Power8 CPU with 12 cores and 96 threads only needs a license for 12 cores to fully utilize the power of the processor.

 

[Nordic]

Uh buying less cores means less overall performance. What is needed is better performance from individuals cores. If enterprises only need to pay for 12 cores to get the same performance as Intel's 18 cores then that's a huge savings license-wise.

So your saying you want even better than what is available. I get that. I want a 6ghz quad core for my desktop too.

 

[IvantheDugtrio]

So your saying you want even better than what is available. I get that. I want a 6ghz quad core for my desktop too.

Well in IBM's case they just have a faster CPU with fewer cores than Intel. Also their CPUs are clocked between 3.0 GHz and 5.0 GHz. Trouble is it can't run Crysis .

Intel's just sitting idle and adding more cores doesn't really improve upon the last generation.

 

[Prima.Vera]

Well in IBM's case they just have a faster CPU with fewer cores than Intel. Also their CPUs are clocked between 3.0 GHz and 5.0 GHz. Trouble is it can't run Crysis .

Intel's just sitting idle and adding more cores doesn't really improve upon the last generation.

What's the power consumption on those 5Ghz CPUs?

 

[IvantheDugtrio]

What's the power consumption on those 5Ghz CPUs?

Power consumption isn't explicitly stated but the 12-core CPUs have a 195W TDP plus 20W TDP for each memory buffer (8 per socket).

 

[Nordic]

I haven't heard about ibm's server chips. They are pretty interesting. I just read this. It isn't x86 though which makes it not as widely applicable. Looks like a good chip.

 

[twilyth]

That's a great article. What I don't really understand though is why you couldn't run virtual x86 machines in those threads. I understand that for server farms that would mean a loss of efficiency and so for that purpose it's not an option. But for other applications, why wouldn't that work?

Anyway, the power 8 chips will likely start life as hosts for Watson servers. IBM is investing $1B in porting Watson to the cloud and making it accessible to developers. Since Watson is an IBM product, I assume it was developed on Power chips.

This is perfect vertical integration for IBM assuming that they can maintain the lead that Watson has over other data analytics competitors. They create the software and the hardware it runs on. I mean damn. If they pull this off, it will be just like the good old days again for them.

 

[FordGT90Concept]

What? This has nothing to do with anti-trust laws. Intel isn't monopolizing anything, and simply making a better product that your competition isn't against the law.

It's not Intel's fault that the vast majority of software only takes advantage of 1 to 2 cores/threads, and Intel already sells multiple chips specifically for the home market that can handle 8-16 threads. It's up to the software to take advantage of that.

Also, a home users CAN purchase server/workstation CPUs if they think their home office/game PC need two 16 core/32 thread CPUs (protip: it doesn't) and they have about $7000 to throw away.

And it's not Intel's fault that AMD can't make competitive chips.

Anti-trust is about anti-competition and Intel's pricing scheme is very anti-competition. They more or less match AMD's prices at the low end but the higher you go, the further the discrpency. Intel can generate huge profit margins off of these processors while AMD, because of their process disadvantage, can only scrape by. Intel is effectively driving AMD out of the market one generation at a time. Intel needs to get slapped with a massive fine for this pricing scheme.

Anti-trust is always about the market, not about what people need or want. I think it is a stretch of the imagination to call this stagnation in consumer chips (especially the i7 brand) fair.

 

[Sony Xperia S]

With this:

AMD Could Potentially Get 19B investment

http://www.guru3d.com/news-story/amd-could-potentially-get-9b-investment.html

and

AMD 16nm Opteron and FX Processors with Upto 20 Cores are a Possibility in 2016-2017

http://wccftech.com/amd-16nm-opteron-fx-processors-possibly-upto-20-cores-2016-2017/

and

AMD Working On Something “Crazy” For GDC – Plus New Demo To Make Starswarm Look Primitive

http://wccftech.com/amd-working-crazy-gdc/#ixzz3Rj2LvnMI

I think the future for AMD is all right.

 

[Aquinus]

That's a great article. What I don't really understand though is why you couldn't run virtual x86 machines in those threads. I understand that for server farms that would mean a loss of efficiency and so for that purpose it's not an option. But for other applications, why wouldn't that work?

Anyway, the power 8 chips will likely start life as hosts for Watson servers. IBM is investing $1B in porting Watson to the cloud and making it accessible to developers. Since Watson is an IBM product, I assume it was developed on Power chips.

This is perfect vertical integration for IBM assuming that they can maintain the lead that Watson has over other data analytics competitors. They create the software and the hardware it runs on. I mean damn. If they pull this off, it will be just like the good old days again for them.

That's because most modern VMs run on hardware using visualization extensions to pass through CPU instructions to hardware to make VMs run as fast as the host machine. If you have an X86 server, you're going to be running X86 VMs because visualization extensions allow the VM to run directly on hardware when the instruction set is the same. This does not accelerate other architectures because you can't run other architectures on hardware because... well, it's not there; the instructions are different and it's all handled differently so emulation is required. Even when Apple moved from PowerPC to Intel, Rosetta (which allowed PPC binaries run on Intel; PPC using Altivec and belonging to IBM and being an early POWER variant) was emulated in software.

The simple fact is that if you want to use CPUs like that, you need to run software that is compiled in that instruction set. Emulating ISAs is dead slow which is why you don't use it in server farms. If you need X86, you use X86. If you need POWER, you use POWER, if you need SPARC, you use SPARC. It's really as simple as that. The only time you really should emulate ISAs is if you're a developer and don't have access to the real hardware it will run on and you need to develop locally. The loss in performance to emulate will always be too much for real world applications in production; the translation is too costly versus running natively.

Also consider the cost of IBM servers plus licensing costs. X86 more often than not is the better option unless you have some very specific uses for a server farm outside of traditional software.

With this:

AMD Could Potentially Get 19B investment

http://www.guru3d.com/news-story/amd-could-potentially-get-9b-investment.html

and

AMD 16nm Opteron and FX Processors with Upto 20 Cores are a Possibility in 2016-2017

http://wccftech.com/amd-16nm-opteron-fx-processors-possibly-upto-20-cores-2016-2017/

and

AMD Working On Something “Crazy” For GDC – Plus New Demo To Make Starswarm Look Primitive

http://wccftech.com/amd-working-crazy-gdc/#ixzz3Rj2LvnMI

I think the future for AMD is all right.

You can't sell news articles and I've been hearing about a lot of good things for a long time that have never truly lived up to their expectations. So in other words, I'll believe it when I see it but until then, you can't sell an unfinished product.

 

[Sony Xperia S]

You can't sell news articles and I've been hearing about a lot of good things for a long time that have never truly lived up to their expectations. So in other words, I'll believe it when I see it but until then, you can't sell an unfinished product.

I am interested. What won't happen? Releasing new processors in 2016-2017, those possible investments poured into AMD or the "crazy" thing?

 

[Aquinus]

I am interested. What won't happen? Releasing new processors in 2016-2017, those possible investments poured into AMD or the "crazy" thing?

I'm sure they'll release new CPUs, the question is how better will they be over what's out now. More often than not, the improvement is overstated. I really don't care if we get 20 core CPUs because no one needs it and only few applications can take advantage of it due to the kinds of workloads that get accelerated by more processes and/or threads with respect to the consumer market.

As far as Opterons are concerned, I can believe a 20c chip, since they've been producing 16c chips for a while now (essentially two 8c dies shoved together,) so I don't doubt it. As for investment, I don't gamble on such things. If it happens we very well might see progress in some specific area. I'm not interested because it doesn't really mean anything as it really just feels like more PR, something AMD has been doing a lot lately without too much to show for it. I don't mean to bash AMD but, they've been playing the PR card a lot lately and it hasn't resulted in anything really worthwhile yet. So I'm reluctant to think things are going to change without seeing these things actually occur.

I hate to say that my confidence in AMD is at an all time low. I take no pleasure in saying that.

 

[Sony Xperia S]

I hate to say that my confidence in AMD is at an all time low. I take no pleasure in saying that.

If so, then don't do it. Don't be so negative, pessimistic, reserved and distant.

Instead, try to focus on how you can help. Of course, if you are honestly willing to...

 

[Aquinus]

If so, then don't do it. Don't be so negative, pessimistic, reserved and distant.

Instead, try to focus on how you can help. Of course, if you are honestly willing to...

Don't tell me how I should feel. My simple point is that I've had confidence in AMD for a very long time and it's been diminishing for just as long. You might not want me to be "so negative, pessimistic, reserved and distant" but how I feel now is a result of years and year of the same old crap. There are more tends to show AMD will continue to do nothing substantial than that it will charge ahead with great technology to win back the market. So I apologize is my negativity is disturbing your zen but I wouldn't need to be negative if AMD hadn't got here in the first place.

Now, to actually shift to the topic instead of bantering about the opposite camp.

Everyone does know that there is a 18c/36t Xeon E5 part for skt2011-3, right? When it actually shows up in the market is another story though.

 

[twilyth]

A pair of 14 cores sold on ebay today for $1200 each but single chips have sold for less than 1k in the recent past. Haven't see the 18c versions show up yet.

 

[Sony Xperia S]

There are more tends to show AMD will continue to do nothing substantial than that it will charge ahead with great technology to win back the market.

If Intel declared a war with all means (which means legal and not legal, fair and not fair) against AMD and you and others like you help Intel in this battle, then indeed nothing substantial would happen.

 

[Aquinus]

If Intel declared a war with all means (which means legal and not legal, fair and not fair) against AMD and you and others like you help Intel in this battle, then indeed nothing substantial would happen.

If you mean by the decision of my purchase? At the time I had 3 decisions, a 3820, 2700k, or a FX 8150. Considering the performance advantages of the 3820 over the 8120 and the amount of expandability skt2011 offered, I didn't really have much option because I chose the best option for my price point. If I wasn't spending almost 2k USD on my last upgrade, it probably wouldn't have been Intel.

So I will say again

Don't tell me how I should feel.

I went Intel because skt2011 provided what I wanted at the time. I buy what I think is best for what I need or am going to need and you shouldn't assume that my lack of faith in AMD means that I wholeheartedly support Intel, because I don't. Although like any consumer, I want the best product and Intel had what I wanted.

Once again, I've said before that we should stay on topic and I wouldn't call this on topic. Let's try to restrict the discussion on Intel's lineup and keep AMD out of it.

 

[Frick]

and keep AMD out of it.

Yeah I wondered what happened to the VIA Isiah II. It sounded good.

EDIT: Seriously though, I find this a bit boring. The interesting things in CPU space happens in the low power market IMO.

 

[RichF]

it's not Intel's fault that AMD can't make competitive chips.

I guess you missed out on the many interesting tidbits in the history of Intel abusing its monopoly power.

Sabotaging its compiler to reduce the performance of AMD processors is just one of those.

Sure... Intel doesn't have anything to do with AMD's poor competitiveness right now... I have some lovely property in Florida you may be interested in.