So I'm surfing the net, and I read this:

http://tinyurl.com/lls9hh

So there are hex-core Operons now, eh? And HP are planning to ship them?

Obviously, I immediately wondered where you buy an 8-socket HP server 
populated with hex-core Operons, and what such a beast would cost.

http://tinyurl.com/mfpfxu

So we don't know yet. However, it's little brother is listed "starting 
at" $16,999.00. (As far as I can tell, the principle difference is that 
the G5 takes quad-core Operons, while the G6 does hex-core.) So I think 
we can safely assume this is out of my price range.



In other news... Some of HP's other servers are actually not as 
expensive as you'd think, considering the horsepower.

Post a reply to this message

Invisible wrote:

> In other news... Some of HP's other servers are actually not as 
> expensive as you'd think, considering the horsepower.

...for example:

http://tinyurl.com/lrkzto

So for $1,500 (roughly £1,000) I get a dual hex-Operon rack-mount 
server. Not too shabby, eh?

Of course, this is quite silly. If I buy a desktop system instead, I can 
probably get similarly impressive CPU performance *and* have somewhere 
to put my expensive GPU. ;-) But, for reasons unknown, desktop 
motherboards never support multiple CPUs...

Post a reply to this message

Invisible <voi### [at] devnull> wrote:
> So there are hex-core Operons now, eh?

  I'm wondering: What do people do with so many cores? And I'm talking about
average people, not people who use POV-Ray 3.7 (who are a rather small
minority).

  About the only usage for multiple cores for the masses (rather than for
very small niche markets) is computer games which support multithreading.

  However, I wonder if anyone has actually benchmarked how well current
multithreading-supporting games scale up with increased amount of cores
(and processors). Expecting that doubling the number of cores will double
the refresh rate of the game is completely unrealistic wishful thinking.
I would be curious to know if anyone has actually benchmarked how much
games *really* benefit from additional cores.

  What I'm getting at is that if, let's say, increasing the amount of
cores from 2 to 6 only about doubles the refresh rate of the game, that's
kind of wasted resources and wasted money. It's like having 2 cores being
idle all the time.

  And of course there's a point after which a game will not benefit at
all from increased speed. A game refreshing at 100FPS or at 150FPS makes
absolutely no practical difference.

  So what do people need so many cores for? Not everyone uses them for
multithreaded rendering or video editing.

-- 
                                                          - Warp

Post a reply to this message

Warp wrote:
> Invisible <voi### [at] devnull> wrote:
>> So there are hex-core Operons now, eh?
> 
>   I'm wondering: What do people do with so many cores? And I'm talking about
> average people, not people who use POV-Ray 3.7 (who are a rather small
> minority).
> 
<snip>
>   So what do people need so many cores for? Not everyone uses them for
> multithreaded rendering or video editing.
> 
Who cares what they use them for? The more people buying them, the lower
the price gets to you and me :-)

<Marketing> C'mon suckers, buy your redundant cores now. Be the first on
your block to watch YouTube on our new dodecacore Hyperon! </Marketing>

John
-- 
"Eppur si muove" - Galileo Galilei

Post a reply to this message

Warp wrote:

>   I'm wondering: What do people do with so many cores? And I'm talking about
> average people, not people who use POV-Ray 3.7 (who are a rather small
> minority).

Indeed, where I work we have all these shiny new Dell OptiPlex 755 
systems with Intel Core 2 Duo 2.2 GHz CPUs... and we use them to run 
Word and Excel.

Over 10 years ago, I was running Word and Excel with 486 and Pentium I 
processors. That *was* actually kind of underpowered. But we passed the 
point of being underpowered quite some time ago. Excel isn't notably 
faster on a 2.2 GHz Core 2 Duo than it is on the dinosaur AMD AthlonXP 
1700+ at my desk. And that's a single-core, 32-bit CPU.

Most of the PCs round here, it seems, are in fact limited by RAM 
capacity, HD access speed, and most significantly network latency. 
Downloading a 200 MB Word document takes a while no matter what CPU you 
have.

>   About the only usage for multiple cores for the masses (rather than for
> very small niche markets) is computer games which support multithreading.

Yeah, just about.

>   However, I wonder if anyone has actually benchmarked how well current
> multithreading-supporting games scale up with increased amount of cores
> (and processors).

I think Tom's Hardware has taken a look at this kind of thing on 
occasion, yes.

> Expecting that doubling the number of cores will double
> the refresh rate of the game is completely unrealistic wishful thinking.

Yes, "most" games are GPU-limited.

> I would be curious to know if anyone has actually benchmarked how much
> games *really* benefit from additional cores.

I don't have the benchmarks to hand, but I'm fairly sure Tom's Hardware 
has looked at this once or twice. (E.g., I vaguely recall an article 
about whether you should upgrade your CPU or your GPU, and which would 
give you the most improvement for the same money.)

>   And of course there's a point after which a game will not benefit at
> all from increased speed. A game refreshing at 100FPS or at 150FPS makes
> absolutely no practical difference.

Ah, but you're forgetting the first rule of 3D graphics: If you add more 
processing power, people will invent new ways to sap it until the thing 
slows to a crawl.

Exhibit A: Crysis.

Post a reply to this message

Doctor John wrote:

> Who cares what they use them for? The more people buying them, the lower
> the price gets to you and me :-)

Bahaha! Dude, that's evil.

Post a reply to this message

Invisible wrote:
> Doctor John wrote:
> 
>> Who cares what they use them for? The more people buying them, the lower
>> the price gets to you and me :-)
> 
> Bahaha! Dude, that's evil.

Evil, moi?
Twisted, maybe, but evil, never }:->

John
-- 
"Eppur si muove" - Galileo Galilei

Post a reply to this message

Warp schrieb:
>   I'm wondering: What do people do with so many cores? And I'm talking about
> average people, not people who use POV-Ray 3.7 (who are a rather small
> minority).

Damn, and I was already about to hit the "reply" button to answer that 
perfectly obvious question :-)

Well, so what might /others/ do with such fast computers...

Brag about them?
Hope for games to support them at last?

Can't think of much else.

>   And of course there's a point after which a game will not benefit at
> all from increased speed. A game refreshing at 100FPS or at 150FPS makes
> absolutely no practical difference.

Forget about that: Computer games are like POV-Ray scenes - they 
invariably grow with the available computing power, to bog it down again 
to exactly that speed where it's /barely/ acceptable...

There's always stuff to improve you know. If they'd find nothing else, 
they'd go for realtime subsurface scattering and crash-test-quality 
physics simulation...

Post a reply to this message

>  I'm wondering: What do people do with so many cores? And I'm talking 
> about
> average people, not people who use POV-Ray 3.7 (who are a rather small
> minority).

Nothing, they just buy them because it sounds better than before (after all, 
they can't make the clock rate much higher anymore).  A bit like how a phone 
with a 5MP camera sells better than one with a 2MP camera, even though the 
photos are just as crappy and you can hardly blow it up to A4 to print out.

Post a reply to this message

Doctor John wrote:

> 
> <Marketing> C'mon suckers, buy your redundant cores now. Be the first on
> your block to watch YouTube on our new dodecacore Hyperon! </Marketing>

I want one!

-- 
~Mike

Post a reply to this message

>>  I'm wondering: What do people do with so many cores? And I'm talking 
>> about
>> average people, not people who use POV-Ray 3.7 (who are a rather small
>> minority).
> 
> Nothing, they just buy them because it sounds better than before (after 
> all, they can't make the clock rate much higher anymore).  A bit like 
> how a phone with a 5MP camera sells better than one with a 2MP camera, 
> even though the photos are just as crappy and you can hardly blow it up 
> to A4 to print out.

It's completely true! My mum bought a 7MP camera and was bragging about 
how much better it was than my little 3MP camera...

...until we looked at the actual pictures, and discovered that mine was 
vastly superior. Funny how having a 0.2 mm lense results in awful 
picture quality, eh?

For what most people do, the single limiting factor in apparent speed is 
the HD and the Internet link. But while you could print a page of stuff 
about rotational latency and cache size, it's easier to slap a big "Core 
i7 3.0 GHz" sticker on the front instead.

Big numbers sell.

Post a reply to this message

clipka <ano### [at] anonymousorg> wrote:
> Forget about that: Computer games are like POV-Ray scenes - they 
> invariably grow with the available computing power, to bog it down again 
> to exactly that speed where it's /barely/ acceptable...

  But the thing is: Regardless of how many cores you add to your computer,
the games will quickly reach a roof where they won't get any faster.

  I'm willing to bet that you won't see any significant improvement in
framerates between a 2-core and a 6-core computer, which otherwise has
identical hardware.

  The main reason is that the GPU sets the limit, not the CPU. If the CPU
gets fast enough, it will just sit idle while the GPU renders a frame.
Adding more cores is not going to help that.

  Another reason is that, as I said, the computational capacity of the
game engine will most probably not grow linearly with the number of cores.
Performance which is completely linear compared to the number of cores is
a really hard problem. I would bet that with a game engine adding one more
core is probably only going to speed up calculations by 50% or something
like that.

  If you want more speed, the cores themselves would need to be faster.

-- 
                                                          - Warp

Post a reply to this message

Invisible wrote:
> But, for reasons unknown, desktop motherboards never support multiple CPUs...

I dunno. I go down to Frye's Electronics and they have a bunch. They're all 
hundreds of dollars more than the single-CPU boards...

-- 
   Darren New, San Diego CA, USA (PST)
   Understanding the structure of the universe
    via religion is like understanding the
     structure of computers via Tron.

Post a reply to this message

Warp wrote:
>   So what do people need so many cores for? Not everyone uses them for
> multithreaded rendering or video editing.

Photo editting. Video encoding. Background stuff like virus scans while 
you're trying to get actual work done. :-)

And server-type stuff: compiling, SQL servers, web servers, virtual 
machines, etc.

I often wind up using at least two cores, and not infrequently three.
Only while video encoding does something use more than three, altho I 
haven't had a distributed compile so big I could see it using up all the 
cores long enough for it to show.

-- 
   Darren New, San Diego CA, USA (PST)
   Understanding the structure of the universe
    via religion is like understanding the
     structure of computers via Tron.

Post a reply to this message

Invisible wrote:
> Most of the PCs round here, it seems, are in fact limited by RAM 
> capacity, HD access speed, and most significantly network latency. 

My boss bought a quad-core Mac with SSDs. A few weeks later I asked how it 
worked out.  He said "I never wait for anything." :-)

With several SATA drives, it's nice to be able to copy at 80MBps between two 
different pairs of drives at once.  Way nicer than lame-ass IDE.

And my net here is nicely peppy. At 12Mbps, I'm almost always maxing out 
someone else's connection before my own. I can suck down an entire CD worth 
of data in about 15 minutes, faster than driving into work to pick it up.

-- 
   Darren New, San Diego CA, USA (PST)
   Understanding the structure of the universe
    via religion is like understanding the
     structure of computers via Tron.

Post a reply to this message

Darren New <dne### [at] sanrrcom> wrote:
> Warp wrote:
> >   So what do people need so many cores for? Not everyone uses them for
> > multithreaded rendering or video editing.

> Photo editting.

  Does the average computer user both own a photo editing software which
supports multithreading and use it to its full extent?

> Video encoding.

  Niche market, i'd say. Even those who do video encoding could probably
do it just as well with a dual-core or even a single-core. The pros who
really need the cores are mostly a niche market (and they also tend to buy
high-end computers because they need TONS of memory and MEGATONS of hard
drive space, not to talk about all the backup storage space. Not the
average joe's computer.)

> Background stuff like virus scans while 
> you're trying to get actual work done. :-)

  Virus scanners mostly stress disk I/O, which is the main reason why
everything's so laggy during a scan. From the CPU point of view a modern
single-core one handles it just fine, with other processes running at
the same time. Multiple cores (especially more than two) are not going
to make any difference.

> And server-type stuff: compiling, SQL servers, web servers, virtual 
> machines, etc.

  Most servers work just fine with single-core CPUs, especially if they
are not shared servers (as most servers out there aren't). High-end servers
with many customers may be a different story, but also a niche market
(compared to the *average* user). Those, too, tend to have higher specs
than the average PC.

> I often wind up using at least two cores, and not infrequently three.
> Only while video encoding does something use more than three, altho I 
> haven't had a distributed compile so big I could see it using up all the 
> cores long enough for it to show.

  You are a power user, not an average user. And even you admit that you
seldom need even three cores. More is just a waste.

  Well, except when rendering something with povray... :P

-- 
                                                          - Warp

Post a reply to this message

Warp wrote:
>   Does the average computer user both own a photo editing software which
> supports multithreading and use it to its full extent?

I dunno. Who is an average programmer? :-)  If you do a "sharpen" or 
something and it's written to use multiple threads to do so, you'll use all 
the cores.

>   Niche market, i'd say. 

Sure. Almost everything in computers is a niche market in some sense. :-)

> Even those who do video encoding could probably
> do it just as well with a dual-core or even a single-core.

Of course. Just not as fast. Everyone who downloads videos in the first 
place could do it over dialup, too. (Except these are my own videos I'm 
transcoding, of course)

>> Background stuff like virus scans while 
>> you're trying to get actual work done. :-)
> 
>   Virus scanners mostly stress disk I/O,

Sure. I just meant like other background stuff one tends to accumulate.

>   You are a power user, not an average user. And even you admit that you
> seldom need even three cores. More is just a waste.

Agreed. I was just giving samples of why one might want more cores that you 
might not have thought of.

-- 
   Darren New, San Diego CA, USA (PST)
   Understanding the structure of the universe
    via religion is like understanding the
     structure of computers via Tron.

Post a reply to this message

Warp wrote:
> Invisible <voi### [at] devnull> wrote:
>> So there are hex-core Operons now, eh?
> 
>   I'm wondering: What do people do with so many cores? And I'm talking
>   about
> average people, not people who use POV-Ray 3.7 (who are a rather small
> minority).
> 
> [...]
> 
>   So what do people need so many cores for? Not everyone uses them for
> multithreaded rendering or video editing.

Many web browsers are now becoming multithreaded or using multiple
processes. So now you'll be able to use browser tabs, each with some
bloated Javascript monstrosity, without any of them slowing down!

Post a reply to this message

Darren New <dne### [at] sanrrcom> wrote:
> > Even those who do video encoding could probably
> > do it just as well with a dual-core or even a single-core.

> Of course. Just not as fast.

  Fast enough, in most cases. I think memory and I/O speed will be the
bottleneck at some point, after which the extra cores will be worth a
paperweight.

  (It's surprising how much effect memory bus speed has eg. on video
capturing.)

-- 
                                                          - Warp

Post a reply to this message

Nicolas Alvarez <nic### [at] gmailcom> wrote:
> Many web browsers are now becoming multithreaded or using multiple
> processes. So now you'll be able to use browser tabs, each with some
> bloated Javascript monstrosity, without any of them slowing down!

  I'd call that a complete waste of resources and money. :)

-- 
                                                          - Warp

Post a reply to this message

Warp wrote:

>   Fast enough, in most cases. I think memory and I/O speed will be the
> bottleneck at some point, after which the extra cores will be worth a
> paperweight.
> 
>   (It's surprising how much effect memory bus speed has eg. on video
> capturing.)

And this is the other Fun Thing. Given enough CPU cores, you will 
eventually reach a point where the memory subsystem can't actually keep 
up. The result is that the more cores you have, the more time they can 
waste sitting idle.

(As somebody once said, "a supercomputer is a device for turning 
compute-bound problems into I/O-bound problems". It seems apt here.)

Long ago when RAM was actually faster than the CPU, having several of 
them seemed like a good idea. Currently the CPU goes way, way faster 
than RAM anyway. Having "more CPU" just makes the problem worse...

-- 
http://blog.orphi.me.uk/
http://www.zazzle.com/MathematicalOrchid*

Post a reply to this message

Orchid XP v8 <voi### [at] devnull> wrote:
> And this is the other Fun Thing. Given enough CPU cores, you will 
> eventually reach a point where the memory subsystem can't actually keep 
> up. The result is that the more cores you have, the more time they can 
> waste sitting idle.

  In theory you could still get an advantage if each core has its own
L1 cache and they perform heavy calculations on bunches of data which
fit those caches (and likewise the routine itself must obviously also
fit in the L1 cache).

  In other words, if the algorithm can be constructed to be of the type
"crunch 4 kB of data for several seconds, write the results to RAM and
read new 4 kB of data, repeat", then additional cores will give an
advantage.

  Of course many algorithms deal with a lot more data at a time than will
nicely fit in L1 cache, especially if the cache is shared among the cores,
so efficiency problems will start happening.

  (Ironically, with shared L1 cache systems you might end up in situations
where a single-threaded version of the algorithm actually runs faster than
a multithreaded version, or where the multithreaded one doesn't run any
faster than the single-threaded one.)

-- 
                                                          - Warp

Post a reply to this message

Warp schrieb:
>   The main reason is that the GPU sets the limit, not the CPU. If the CPU
> gets fast enough, it will just sit idle while the GPU renders a frame.
> Adding more cores is not going to help that.

That's only part of the story.

The other part is "AI", physics sim and stuff like that. If you've maxed 
out the GPU and have no way to add visual incentives to buy, you can 
start focusing on making the gameplay more complex.

If the gaming industry won't find any other way to keep additional CPU 
cores busy, they'll go on smartening the AIs, adding more physics 
effects, getting rid of level transitions, and what-have-you-not.

And maybe ultimately they might even start inventing innovative game 
concepts ;-)

Post a reply to this message

>  The main reason is that the GPU sets the limit, not the CPU. If the CPU
> gets fast enough, it will just sit idle while the GPU renders a frame.
> Adding more cores is not going to help that.

More realistic physics and AI!

Post a reply to this message

>> But, for reasons unknown, desktop motherboards never support multiple 
>> CPUs...
> 
> I dunno. I go down to Frye's Electronics and they have a bunch. They're 
> all hundreds of dollars more than the single-CPU boards...

Yeah, I imagine it's pretty expensive to wire up several hundred extra 
tracks on the board. SLI boards are all way more expensive too. But at 
least shops *sell* those...

Post a reply to this message

Darren New wrote:

> My boss bought a quad-core Mac with SSDs. A few weeks later I asked how 
> it worked out.  He said "I never wait for anything." :-)

And that's the difference. With Windoze, just closing the CD drive is 
enough to lock the entire Explorer shell for ten minutes while it 
attempts to determine whether there's a disk in there. (Um, 
multitasking? Anyone?)

> With several SATA drives, it's nice to be able to copy at 80MBps between 
> two different pairs of drives at once.  Way nicer than lame-ass IDE.

My PC at home is all SATA too. It still takes forever for TF2 to start 
up. :-P

> And my net here is nicely peppy. At 12Mbps, I'm almost always maxing out 
> someone else's connection before my own. I can suck down an entire CD 
> worth of data in about 15 minutes, faster than driving into work to pick 
> it up.

Where in the name of God can you get 12 Mbit/sec? I thought 8 was the 
maximum that ADSL supports...

Post a reply to this message

>> And this is the other Fun Thing. Given enough CPU cores, you will 
>> eventually reach a point where the memory subsystem can't actually keep 
>> up. The result is that the more cores you have, the more time they can 
>> waste sitting idle.
> 
>   In theory you could still get an advantage if each core has its own
> L1 cache and they perform heavy calculations on bunches of data which
> fit those caches (and likewise the routine itself must obviously also
> fit in the L1 cache).
> 
>   In other words, if the algorithm can be constructed to be of the type
> "crunch 4 kB of data for several seconds, write the results to RAM and
> read new 4 kB of data, repeat", then additional cores will give an
> advantage.
> 
>   Of course many algorithms deal with a lot more data at a time than will
> nicely fit in L1 cache, especially if the cache is shared among the cores,
> so efficiency problems will start happening.

Typically the L1 cache is per-core, and the L2 cache is shared. But the 
point still stands: If two cores try to write to the same region of 
memory, they tend to constantly trip other each other with cache 
coherancy issues. Oh, and if your algorithm needs random access to a 
large block of RAM, forget it.

>   (Ironically, with shared L1 cache systems you might end up in situations
> where a single-threaded version of the algorithm actually runs faster than
> a multithreaded version, or where the multithreaded one doesn't run any
> faster than the single-threaded one.)

This is quite common. I've been reading documentation about Haskell's GC 
engine. They found that when running in parallel, sometimes it's faster 
to turn *off* load-balancing, because that way each GC thread is 
processing data which already happens to be in the cache, so that's 
faster. If you migrate the work to another core, the data has to be 
pumped out of one cache, into memory, and into the other cache before it 
can be processed.

If CPUs didn't need caches in the first place (i.e., RAM was faster than 
the CPU) then this would be a total non-issue. But here in the real 
world, it's sometimes faster to leave cores idling rather than risk 
upsetting the Almighty Cache. How sad...

Post a reply to this message

> If CPUs didn't need caches in the first place (i.e., RAM was faster than 
> the CPU) then this would be a total non-issue. But here in the real world, 
> it's sometimes faster to leave cores idling rather than risk upsetting the 
> Almighty Cache. How sad...

Yes, if we upset the Almight Cache, *shock* we might drop back to the 
performance levels of the fastest RAM available.  The cache is there to 
*speed up* stuff, I have no idea why you'd want a machine with a CPU running 
at the same speed as the fastest RAM available, you're then going to get the 
same levels of performance as if you upset the cache continuously!

Post a reply to this message

scott wrote:
>> If CPUs didn't need caches in the first place (i.e., RAM was faster 
>> than the CPU) then this would be a total non-issue. But here in the 
>> real world, it's sometimes faster to leave cores idling rather than 
>> risk upsetting the Almighty Cache. How sad...
> 
> Yes, if we upset the Almight Cache, *shock* we might drop back to the 
> performance levels of the fastest RAM available.

I meant that it's sad that we don't have RAM that can perform as fast as 
the CPU itself.

[Or rather... I guess we do, since that must be what they make the L1 
cache out of. But the L1 cache is tiny, so...]

Post a reply to this message

> I meant that it's sad that we don't have RAM that can perform as fast as 
> the CPU itself.
>
> [Or rather... I guess we do, since that must be what they make the L1 
> cache out of. But the L1 cache is tiny, so...]

And, most importantly, it is very close to the CPU.

It's a cost/benefit thing, for $X how do you make the fastest computer?  The 
answer is to have a big slab of slow RAM, and progressively smaller bits of 
faster RAM.  Trying to do it another way will not make the fastest machine 
for a given amount of money.

Post a reply to this message

>> I meant that it's sad that we don't have RAM that can perform as fast 
>> as the CPU itself.
>>
>> [Or rather... I guess we do, since that must be what they make the L1 
>> cache out of. But the L1 cache is tiny, so...]
> 
> And, most importantly, it is very close to the CPU.
> 
> It's a cost/benefit thing, for $X how do you make the fastest computer?  
> The answer is to have a big slab of slow RAM, and progressively smaller 
> bits of faster RAM.  Trying to do it another way will not make the 
> fastest machine for a given amount of money.

It just seems slightly crazy having a CPU that's hundreds of times 
faster than RAM, and using caching is a trick to try to hide this 
inescapable fact.

I wonder... if money was no object, *could* you actually have a 1GB L1 
cache?? Or is it actually not possible for some other reason?

Post a reply to this message

> It just seems slightly crazy having a CPU that's hundreds of times faster 
> than RAM, and using caching is a trick to try to hide this inescapable 
> fact.

It's not crazy, it's a clever idea to get more performance for less money.

> I wonder... if money was no object, *could* you actually have a 1GB L1 
> cache?? Or is it actually not possible for some other reason?

I'm no expert, but the length of the tracks from the farthest away bit of L1 
cache to the CPU might be a problem.

Post a reply to this message

scott schrieb:
> Yes, if we upset the Almight Cache, *shock* we might drop back to the 
> performance levels of the fastest RAM available.  The cache is there to 
> *speed up* stuff, I have no idea why you'd want a machine with a CPU 
> running at the same speed as the fastest RAM available, you're then 
> going to get the same levels of performance as if you upset the cache 
> continuously!

Just as a side note: Cache technically /is/ RAM.

Not "main memory" though.

(Nor is it typically DRAM, but SRAM instead: Way faster, and way more 
expensive...)

Post a reply to this message

scott schrieb:
> I'm no expert, but the length of the tracks from the farthest away bit 
> of L1 cache to the CPU might be a problem.

If money would be no issue, maybe "stacking" dice would be an option to 
keep lines short.

Then again, there's that heat transfer problem...

Post a reply to this message

> Then again, there's that heat transfer problem...

Hmm yes, given that fast RAM chips on some GPUs need heatsinks, having a 
whole GB at L1 speeds is probably going to cook itself instantly.

Post a reply to this message

Invisible schrieb:
> It just seems slightly crazy having a CPU that's hundreds of times 
> faster than RAM, and using caching is a trick to try to hide this 
> inescapable fact.

Look at it this way: /Despite/ your slow DRAM main memory, you can still 
/benefit/ from a madly fast CPU core. (Or multiple of them, for that 
matter.) Isn't that a great thing actually?

Post a reply to this message

scott schrieb:
>> Then again, there's that heat transfer problem...
> 
> Hmm yes, given that fast RAM chips on some GPUs need heatsinks, having a 
> whole GB at L1 speeds is probably going to cook itself instantly.

Then again, maybe not. SRAMs - which AFAIK are the only type of memory 
that would do that job - are freakishly huge compared to DRAMs, so for 
the same amount of memory you'd need about 30 times the die area - at 
the benefit of also having 30 times better heat transfer.

Furthermore, SRAM cells consume virtually no power as long as they are 
idle, and read access costs only the power to drive the address select 
and output lines. DRAM cells, on the other hand, need additional power 
to perform a full rewrite after each read access - and they need to be 
refreshed frequently, which costs power as well.

I'm not sure how much of a difference it makes, but I imagine SRAM to 
consume quite a good deal less power on the whole.


Still - given that caches in modern CPUs appear to occupy something like 
50% or more of the die area, and that's just something like 10 MB, you'd 
have to stack something like 50 chips on top of each other for 1 GB... I 
don't think there's any way to bring 1 GB of memory close enough to the 
CPU to access it at L1 or even L2 speed.

Post a reply to this message

clipka <ano### [at] anonymousorg> wrote:
> Warp schrieb:
> >   The main reason is that the GPU sets the limit, not the CPU. If the CPU
> > gets fast enough, it will just sit idle while the GPU renders a frame.
> > Adding more cores is not going to help that.

> That's only part of the story.

> The other part is "AI", physics sim and stuff like that. If you've maxed 
> out the GPU and have no way to add visual incentives to buy, you can 
> start focusing on making the gameplay more complex.

  Except that you can't. Not really.

  AI and physics is not really something which you can tune up or down
depending on your CPU speed. They are not like graphics: While graphical
quality can be tuned up and down without affecting gameplay in any way,
AI and physics cannot: If you change them, you are effectively changing
the gameplay. Things won't behave in the same way anymore. It might affect
the difficulty of the game (something which might be cool, but not I don't
anyone would be happy with "you can't play this game at hard difficulty
because your CPU is too slow") or even break it completely (eg. if some
physics simulation is too poor, some puzzle might become unsolvable, or
whatever).

  When you create a game, you have to set its AI and physics quality,
and then it will be settled in stone. If the CPU is so fast that it can
calculate them easily for each frame, adding even more cores is not going
to change anything.

  In fact, even if a faster GPU would allow for a higher framerate, that
would still not affect the quality of the AI and physics. It's not like
the CPU would be instructed to calculate more if the framerate goes up
(because, as said, that would change the gameplay). The results will simply
be interpolated more frequently.

  If you were playing a computer chess game, that would be a completely
different story. However, regular computer games are not like that. They
are rather deterministic (ie. what happens doesn't depend on the speed
of your CPU).

  One thing which a faster CPU can be useful for is for calculating additional
visual effects which cannot be calculated by the GPU. In other words, if
your CPU is fast enough, you could turn on certain visual effects (which
require more CPU power to show). OTOH, nowadays more and more of the visual
tasks are submitted to the GPU, so the CPU has less and less to do per frame.

-- 
                                                          - Warp

Post a reply to this message

scott <sco### [at] scottcom> wrote:
> >  The main reason is that the GPU sets the limit, not the CPU. If the CPU
> > gets fast enough, it will just sit idle while the GPU renders a frame.
> > Adding more cores is not going to help that.

> More realistic physics and AI!

  If you tune down the physics and AI because the CPU is not fast enough
to calculate them, you will be drastically changing gameplay. It would not
be just a cosmetic change. I have never heard of any game doing that
(except for games where AI is the whole idea, eg. computer chess, but
that's a completely different category of games).

  Thus after a certain point, when your current CPU can calculate the
physics and AI just fine, adding more cores won't change anything.

-- 
                                                          - Warp

Post a reply to this message

On Fri, 28 Aug 2009 10:22:53 +0200, Invisible <voi### [at] devnull> wrote:
> Where in the name of God can you get 12 Mbit/sec? I thought 8 was the  
> maximum that ADSL supports...

You really need to start catching up with modern technology.  
Alternatively, you could learn to JFGI.

http://en.wikipedia.org/wiki/Asymmetric_digital_subscriber_line#ADSL_standards



-- 
FE

Post a reply to this message

> Where in the name of God can you get 12 Mbit/sec? I thought 8 was the 
> maximum that ADSL supports...

Had 16 at home for the last 18 months...

Post a reply to this message

scott wrote:
>> Where in the name of God can you get 12 Mbit/sec? I thought 8 was the 
>> maximum that ADSL supports...
> 
> Had 16 at home for the last 18 months...

Interesting. Most of the people I know have 2 Mbit/sec or less.

Post a reply to this message

>> More realistic physics and AI!
>
>  If you tune down the physics and AI because the CPU is not fast enough
> to calculate them, you will be drastically changing gameplay. It would not
> be just a cosmetic change. I have never heard of any game doing that

Sure, but don't you think the minimum system requirements for CPUs in games 
will continue to increase?  Game developers are not going to be satisfied to 
just stick with what they have at the moment.

>  Thus after a certain point, when your current CPU can calculate the
> physics and AI just fine, adding more cores won't change anything.

I can't even begin to imagine any point where a CPU will be able to 
calculate physics perfectly without cutting any corners.  Every increase in 
CPU speed will result in more realistic and believable physics with fewer 
short-cuts.  For starters you need non-ridid bodies for large objects, 
realistically deformable terrain and better collisions.  They all need more 
CPU cycles.

Post a reply to this message

>> Had 16 at home for the last 18 months...
> 
> Interesting. Most of the people I know have 2 Mbit/sec or less.

They are all well below the average of 3.6 then:

http://www.ofcom.org.uk/media/features/brspeeds

Post a reply to this message

scott wrote:
>>> Had 16 at home for the last 18 months...
>>
>> Interesting. Most of the people I know have 2 Mbit/sec or less.
> 
> They are all well below the average of 3.6 then:
> 
> http://www.ofcom.org.uk/media/features/brspeeds

Note: "over 60% of UK consumers currently subscribe to packages 
promising speeds of up to 8 Mbit/s."

So 12 Mbit/sec would be pretty damned fast by comparison.


Either way, it's unclear to me whether this page is talking about 
*connection speed* or *actual traffic throughput*. Having a 2 Mbit 
connectio doesn't actually mean you always receive data that fast...

Post a reply to this message

scott <sco### [at] scottcom> wrote:
> >  Thus after a certain point, when your current CPU can calculate the
> > physics and AI just fine, adding more cores won't change anything.

> I can't even begin to imagine any point where a CPU will be able to 
> calculate physics perfectly without cutting any corners.

  I'm talking about an actual modern game.

  After it easily calculates everything it has to calculate per frame,
adding more cores is not going to speed it up at all.

-- 
                                                          - Warp

Post a reply to this message

Warp schrieb:
>   AI and physics is not really something which you can tune up or down
> depending on your CPU speed.

Stopped reading right here.

I'm perfectly aware of that. But still, game developers will keep 
developing games for high-end computers, and include increasingly 
performance-hungry AI, physics etc.

It's not like the inability to easily tune it up or down would prohibit 
from making it more complex.

And it's not like game developers weren't creative in finding ways of 
scaling /some/ stuff to adjust to different computer performance levels.

Post a reply to this message

Warp schrieb:
>   Thus after a certain point, when your current CPU can calculate the
> physics and AI just fine, adding more cores won't change anything.

... until the next game is released.

Post a reply to this message

Invisible wrote:
> multitasking? Anyone?)

It's multitasking. It's just not the explorer multitasking.

> My PC at home is all SATA too. It still takes forever for TF2 to start 
> up. :-P

How fast do they transfer? I get a smooth 80-100 MBps transfer rate.

> Where in the name of God can you get 12 Mbit/sec? I thought 8 was the 
> maximum that ADSL supports...

Cable.

ADSL will easily go up to the gigabit range if the wire's short enough. 
Anything over about 10 feet and you don't get those speeds, tho. :-)

-- 
   Darren New, San Diego CA, USA (PST)
   Understanding the structure of the universe
    via religion is like understanding the
     structure of computers via Tron.

Post a reply to this message

scott wrote:
> I'm no expert, but the length of the tracks from the farthest away bit 
> of L1 cache to the CPU might be a problem.

That and the number of gates you have to go through to decode an address. 
Every layer of transistors you have to go through to translate an address to 
a cell is another gate time of delay.

-- 
   Darren New, San Diego CA, USA (PST)
   Understanding the structure of the universe
    via religion is like understanding the
     structure of computers via Tron.

Post a reply to this message