I give up being a CPU enthusiast.

[Ferzerp]

Yet you are a person who keeps defending bulldozer on here. How do you reconcile that exactly?

I'm getting the impression that it has more to do with the label on the box than the product itself.

 

[IntelEnthusiast]

Sure the desktop market is down. Overall it is down around 4% but something many of you don't know is that the enthusiast space is up 10% overall. So the enthusiast space is not dead and doesn’t look to be dying any time soon.

Another thing in the not too distant future Intel® Atom™ processors will equal the power consumption of ARM based processors.

Christian Wood
Intel Enthusiast Team

 

[Ferzerp]

The problem is, we used to have games that ate up CPU power. Now all we have is high resolution gaming that we have to just keep throwing more and more video cards at it and in many cases a processor from 3 years ago will suffice.

I think that's what he means when he says not exciting. It's not that I don't do anything with my system either. I can play a game at 2560x1600 on my primary monitor, watch an hd cable feed (via quad cablecard tuner) on my secondary monitor, while serving another channel for xbox 360 as an MC extender, and at the same time recording 1-2 channels (either the same, or different ones).

The only thing I notice is in rare cases the secondary monitor (tv channel), will every now and then get jerky for a few seconds (maybe my next video card upgrade should come with a second, weaker card to drive the second monitor).

edit: technically, I'm using a processor from Q1-10, but you could do the same with a 920 with a more aggressive OC.

 

[Arachnotronic]

Sure the desktop market is down. Overall it is down around 4% but something many of you don't know is that the enthusiast space is up 10% overall. So the enthusiast space is not dead and doesn’t look to be dying any time soon.

Another thing in the not too distant future Intel® Atom™ processors will equal the power consumption of ARM based processors.

Christian Wood
Intel Enthusiast Team

Christian,

I appreciate your input. However, can you shed any more light on the upcoming Atoms (Silvermont in particular) versus the state-of-the-art Cortex A15 chips?

 

[Bryf50]

I think you guys are neglecting the fact that these ARM chips are being used in phones.

If they were to design an ARM CPU for a desktop, it would be an absolute beast. Considering what they are doing now with a 2W TDP, if you multiply that by 60, I have a feeling that we would see very competitive performance compared to what we're using currently in terms of x86 desktop CPUs.

This has been discussed. Theres nothing inherent in the ARM architecture that magically makes it low power. If ARM was to design a chip with the performance of Sandy Bridge(not something they could do any time soon) it would likely be more power hungry than SB because of Intels way superior manufacturing processes. They're basically ahead of everyone else by a few years.

 

[Ferzerp]

This has been discussed. Theres nothing inherent in the ARM architecture that magically makes it low power. If ARM was to design a chip with the performance of Sandy Bridge(not something they could do any time soon) it would likely be more power hungry than SB because of Intels way superior manufacturing processes. They're basically ahead of everyone else by a few years.

ARM fanboys can never grasp this for some reason.

 

[Arachnotronic]

This has been discussed. Theres nothing inherent in the ARM architecture that magically makes it low power. If ARM was to design a chip with the performance of Sandy Bridge(not something they could do any time soon) it would likely be more power hungry than SB because of Intels way superior manufacturing processes. They're basically ahead of everyone else by a few years.

Why can't Intel design an x86 chip that is as power efficient as the ARM stuff on a given process node? Intel DOES have some of the brightest people in the industry...

 

[Idontcare]

Why can't Intel design an x86 chip that is as power efficient as the ARM stuff on a given process node? Intel DOES have some of the brightest people in the industry...

Because they aren't in the business of just developing products that offer me-too capabilities from a business perspective, and that means profitability and margins.

Does it make sense for Intel to deploy their best and brightest to work on a project that if successful will enable them to enter a market to chase after 40-50% gross margins when they could deploy those same talented engineers to work on other CPU projects which enable them to develop products that can command 60-70% gross margins?

When and if Intel enters the ARM marketspace they will only do so once they are convinced they can field a product that is capable of performing such that it can command 60% gross margins. If they can't get there then they'll abort the effort (see Larrabee, see HDTV, see prior mobile cellphone projects). An x86 product that merely performs comparable to ARM in the same power footprint is not going to command more money than an ARM chip, and ARM chips do not command 60% margins.

What you see from Intel isn't representative of what they can do, it is representative of what they are prioritizing from the strategic standpoint of maximizing gross margins while also maximizing revenue.

 

[Pederv]

I gave up being a CPU Enthusiast a long time ago. Now a PC is just a tool to be used to accomplish more imaginative goals.
My next goal is a sub-orbital atmospheric monitoring package (helium balloon with a light-weight computer controlled instrument package).

 

[Idontcare]

I gave up being a CPU Enthusiast a long time ago. Now a PC is just a tool to be used to accomplish more imaginative goals.
My next goal is a sub-orbital atmospheric monitoring package (helium balloon with a light-weight computer controlled instrument package).

I have a sub-orbital atmospheric monitoring package too, it sits on my deck and tells me the temperature

 

[Pederv]

I have a sub-orbital atmospheric monitoring package too, it sits on my deck and tells me the temperature

Nice one! LOL

I stop by Anandtech to check out the articles and read a few forum posts but I've been spending most of my time in astronomy and now getting back into Ham Radio. Combining those two and PC knowledge should lead to some interesting projects.

 

[Seero]

I wish ARM can beat the crap out Intel CPU. After they have won I hope Intel comes back with a better CPU that cracks ARM's balls, and then AMD leaves them with dust with another bad ass APU. I really wish...

 

[frozentundra123456]

Because they aren't in the business of just developing products that offer me-too capabilities from a business perspective, and that means profitability and margins.

Does it make sense for Intel to deploy their best and brightest to work on a project that if successful will enable them to enter a market to chase after 40-50% gross margins when they could deploy those same talented engineers to work on other CPU projects which enable them to develop products that can command 60-70% gross margins?

When and if Intel enters the ARM marketspace they will only do so once they are convinced they can field a product that is capable of performing such that it can command 60% gross margins. If they can't get there then they'll abort the effort (see Larrabee, see HDTV, see prior mobile cellphone projects). An x86 product that merely performs comparable to ARM in the same power footprint is not going to command more money than an ARM chip, and ARM chips do not command 60% margins. What you see from Intel isn't representative of what they can do, it is representative of what they are prioritizing from the strategic standpoint of maximizing gross margins while also maximizing revenue.

I understand what you are saying, but if (and I am not saying it will) ARM encroaches enough into the x86 space, Intel will have no choice but to compete there, even if the margin is smaller. Personally, as I said in another post, I could see this happening in the consumer/entertainment area, but probably not in the business/scientific area for a long time. What I am trying to say is if the competition is strong enough, they will have to compete in different markets at possibly lower margins.

 

[exar333]

Intel is turning the PC business into the 1970s US auto business. They got the big CPU engines, but little else,and the answer for everything is more CPU horsepower.

That is one of the most senseless and materially wrong posts I have read in a long time.

If by releasing new processes and archs on a regular basis, and providing more and more efficiency, thats 'stagnating' so be it.

 

[Idontcare]

I understand what you are saying, but if (and I am not saying it will) ARM encroaches enough into the x86 space, Intel will have no choice but to compete there, even if the margin is smaller. Personally, as I said in another post, I could see this happening in the consumer/entertainment area, but probably not in the business/scientific area for a long time. What I am trying to say is if the competition is strong enough, they will have to compete in different markets at possibly lower margins.

There's a reason Intel is spending so much on R&D to increase the process technology gap as much as they are.

ARM may well attempt to get into the high(er) performance markets, but good luck to them in gaining access to a remotely competitive process node to enable their aspirations. AMD can't even do that and they've nearly died trying.

 

[HeXen]

windows 8 wills support ARM. i can see more laptops using ARM instead. But really though, for things like gaming, you don't need much other than plenty of ram and good GpU. i'm on phenom quad and i easily get 60fps in most games

 

[Sunburn74]

Personally what I've noticed is that video games these days made for PCs are so backwards in their requirements that they negate the gains made by the advances in the cpu and gpu areas. Whilst 6000 series and 500 series radeon and nvidia gpus are out, games still have recommended gpus of 8800gtxs (cards that are at least what 5 generations behind?)

I blame consoles personally for retarding this development. The success of the console market doesn't give developers the incentive to push graphics to where games can only be run with high end PC parts. So you quickly reach a point where you can't argue for that extra bit of cpu firepower or gpu power. It doesnt make sense.

The pc enthusiast space is dead and consoles killed it.

 

[Nemesis 1]

Sure the desktop market is down. Overall it is down around 4% but something many of you don't know is that the enthusiast space is up 10% overall. So the enthusiast space is not dead and doesn’t look to be dying any time soon.

Another thing in the not too distant future Intel® Atom™ processors will equal the power consumption of ARM based processors.

Christian Wood
Intel Enthusiast Team

Thanks for the word .

 

[Nemesis 1]

Nice one! LOL

I stop by Anandtech to check out the articles and read a few forum posts but I've been spending most of my time in astronomy and now getting back into Ham Radio. Combining those two and PC knowledge should lead to some interesting projects.

Please tell us more. I to am interested in Ham Radio and astronomy. I guess thats the only good thing I ever got from the schooling @ HRCC school . Ironic isn't it . The church that pushed the flat earth hype when they new perfectly well it was round.

 

[Nemesis 1]

Christian,

I appreciate your input. However, can you shed any more light on the upcoming Atoms (Silvermont in particular) versus the state-of-the-art Cortex A15 chips?

Leave the guy alone . He has done no hyping in this forum . Keeps it simple and honest he does. Google is your friend .

http://www.anandtech.com/show/4333/intels-silvermont-a-new-atom-architecture

 

[Arachnotronic]

In what way am I "bothering" Mr. Wood? He has very valuable contributions when it comes to Intel related products and I just figured he'd have some more info

 

[Idontcare]

Please tell us more. I to am interested in Ham Radio and astronomy. I guess thats the only good thing I ever got from the schooling @ HRCC school . Ironic isn't it . The church that pushed the flat earth hype when they new perfectly well it was round.

What church pushed the "flat earth hype"?

http://en.wikipedia.org/wiki/Myth_of_the_Flat_Earth

The myth of the Flat Earth is the modern misconception that the prevailing cosmological view during the Middle Ages saw the Earth as flat, instead of spherical.[1]
This idea seems to have been widespread during the first half of the 20th century, so that the Members of the Historical Association in 1945 stated that:

"The idea that educated men at the time of Columbus believed that the earth was flat, and that this belief was one of the obstacles to be overcome by Columbus before he could get his project sanctioned, remains one of the hardiest errors in teaching." [2]
​

During the early Middle Ages, virtually all scholars maintained the spherical viewpoint first expressed by the Ancient Greeks. By the 14th century, belief in a flat earth among the educated was essentially dead.

The notion that a flat-earth religious stereotype existed in the dark or middle ages is actually a creation of "creative" historical concoctions of more recent times which, much like the proponents of debunked cold-fusion, takes more time to convince people to forget than it took them to think they were right in the first place.

 

[Ferzerp]

Ok, what about 6,000 year old earth loonies of the present day?

 

[Idontcare]

Ok, what about 6,000 year old earth loonies of the present day?

I'd argue that we should appreciate the willingness of some individuals to hold steadfast to diverse and unpopular ideas. Even today there is such a thing as the flat-earth society.

I don't agree with the tenets of either group, but at the same time I can't think of anything more damaging and unproductive to the health and future of our modern civilization than the idea that we should all be thinking alike and believing/accepting of the same exact ideas and philosophies.

It is the tenacious stubborn mentality of mankind that preserves our ability to overcome, in time, ideas and schools of thought that turn out to be dead wrong.

The more important fundamental laws and facts of physical science have all been discovered, and these are now so firmly established that the possibility of their ever being supplanted in consequence of new discoveries is exceedingly remote . . . Our future discoveries must be looked for in the sixth place of decimals.
Albert A. Michelson, 1894

There is nothing new to be discovered in physics now. All that remains is more and more precise measurement.
William Thomson (Lord Kelvin), 1900

Quantum physics and the technological advancements of the 20th century would not have come about had a few physicists that went against the prevailing wisdom at the time (referred to as the classical Newtonian physics and Maxwell's equations) not stood their ground in pursuit of establishing a school of thought that came to be referred to as the Copenhagen Interpretation.

Werner Heisenberg had been an assistant to Niels Bohr at his institute in Copenhagen during part of the 1920's, when they helped originate quantum mechanical theory. In 1929, Heisenberg gave a series of invited lectures at the University of Chicago, explaining the new field of quantum mechanics.

I don't subscribe to the tenets of the flat earth society, nor the tenets of the Young Earth creationism (the earth is 6000 yrs-old beliefs), but I do celebrate them for that which they represent - dissenting viewpoints and a stubbornness to accept a mono-view of the nature of physics based on its static snapshot at any given point in time.

 

[RussianSensation]

Not to mention the fact that a GTX 480 using over 150 watts more than it's competitor

In gaming, the difference is about 90W or so. 150W is likely under a power virus like Furmark.

Now that the dust settled and reality of DX11 games set in, the GTX480 is 22% faster on average than the HD5870 and still yet slightly faster than the HD6970 up to 1920x1200.

The equivalent comparison would have an FX-8150 be about 20% faster than 2600k and in 6 months when Ivy Bridge (~ HD6970 refresh equivalent) gets released, the FX-8150 would still be slightly faster.....

The comparison to Fermi makes no sense.

is really an order of magnitude worse than bulldozer using an extra 20-50 watts on heavy load.

You can't use Bulldozer on its own, you need a motherboard.

At stock speeds, the FX-8150 system consumes about 80W more than a 2500k system.

Here is another review that shows a 70W difference for system:

However, most people here overclock their CPUs. That difference grows to an astounding 250-270W between a 5ghz 2500k and a 4.8ghz FX-8150. Problem is, Bulldozer gains a lot less performance in overclocked states than an overclocked 2500k since it has worse IPC and more aggressive Turbo Boost to begin with. On the contrary GTX480 overclocked better than the HD5870 and Fermi also scales better with overclocking.

So in summary:

- Fermi was faster right out of the gate, and that lead only extended in DX11 games to let the original Fermi still be as fast/faster than an HD6970
- Fermi had better overclocking and scaling when overclocked
- Fermi offered extra features such as superior Tessellation, NV's superior 3D gaming surround, CUDA, PhysX
- Fermi also had more VRAM than the HD5870
- Fermi was 6 months late not 10 months late
- Fermi has allowed NV to retain a 59% desktop discrete market share
- GTX460 was one of the best mid-range cards this generation
- GTX570 is easily as good as the more expensive 6970 and GTX580 is single GPU king atm vs. Bulldozer that looks hopeless against a 2500k and has no shot whatsoever of beating Ivy Bridge.

vs.

- Bulldozer has worse performance, and it gains less performance when overclocked than Sandy
- Bulldozer's power consumption difference vs. an overclocked 2500k skyrockets to well over 200W
- Bulldozer will have 0 chance of beating an Ivy Bridge (Intel's refresh)
- Bulldozer offers 0 extra features over Intel's CPUs that anyone can actually use
- Bulldozer also forces you to buy an AM3+ motherboard that has inferior SATA 3 performance.
- Bulldozer has made a less than stellar 1090T for $160 suddenly exciting again for AMD users......that would be the equivalent of a GTX285 being a better buy than a GTX480....

It makes no sense to even begin to compare Bulldozer to Fermi. Bulldozer is a huge failure in every way imaginable. It is a way worse CPU than Phenom I was since it costs MORE than a faster 2500k.