Looking for advice for a new PC....
 

I will probably buy it in the spring.

Requirements:
- Windows XP or Vista(?)
- Intel or AMD or? chipset (i.e. not MAC or -NIX or Other)
- FAST! Processor and FAST! Memory
- Probably at least a quad (or are 8's out and affordable now?)
- Prime use will be intensive computing: i.e. GIMPS and related stuff.
- I am NOT really into games or graphics or audio or video ... not that I won't have an audio/video card; just probably a basic one.
- Don't want to have to get a mortgage for it ... I hope I can get what I need for under $2,000.
- I live in Canada so it needs to be available there.

Advice???

Processor of choice: [url]http://www.tigerdirect.ca/applications/SearchTools/item-details.asp?EdpNo=2643933&Sku=CP2-DUO-Q6600[/url]

[QUOTE=petrw1;120474]I will probably buy it in the spring.

Requirements:
- Windows XP or Vista(?)
- Intel or AMD or? chipset (i.e. not MAC or -NIX or Other)
- FAST! Processor and FAST! Memory
- Probably at least a quad (or are 8's out and affordable now?)
- Prime use will be intensive computing: i.e. GIMPS and related stuff.
- I am NOT really into games or graphics or audio or video ... not that I won't have an audio/video card; just probably a basic one.
- Don't want to have to get a mortgage for it ... I hope I can get what I need for under $2,000.
- I live in Canada so it needs to be available there.

Advice???[/QUOTE]

Avoid Vista. It is a memory pig. There are too many GD processes,
many of which are not needed.

Be wary of quad processors. They won't be much faster than duos
unless your jobs are small.

[QUOTE=petrw1;120474]I will probably buy it in the spring.

Requirements:
- Windows XP or Vista(?)
- Intel or AMD or? chipset (i.e. not MAC or -NIX or Other)
- FAST! Processor and FAST! Memory
- Probably at least a quad (or are 8's out and affordable now?)
- Prime use will be intensive computing: i.e. GIMPS and related stuff.
- I am NOT really into games or graphics or audio or video ... not that I won't have an audio/video card; just probably a basic one.
- Don't want to have to get a mortgage for it ... I hope I can get what I need for under $2,000.
- I live in Canada so it needs to be available there.

Advice???[/QUOTE]

[URL]http://www.newegg.com/Product/Product.aspx?Item=N82E16832116378[/URL]
[URL]http://www.newegg.com/Product/Product.aspx?Item=N82E16819115017[/URL]
[URL]http://www.newegg.com/Product/Product.aspx?Item=N82E16813131237[/URL]
[URL]http://www.newegg.com/Product/Product.aspx?Item=N82E16820145098[/URL]
[URL]http://www.newegg.com/Product/Product.aspx?Item=N82E16822136131[/URL]

With all of that, you get a 64bit operating system, a motherboard, a quad-core processor, 2GB of dual channel RAM, high quality integrated graphics and audio and a 750GB hard drive for $744.95, plus approximately $8.41 shipping and handling. All you need now is a good case and a good psu. Here are my suggestions:

[URL]http://www.newegg.com/Product/Product.aspx?Item=N82E16817104902[/URL]
[URL]http://www.newegg.com/Product/Product.aspx?Item=N82E16811129154[/URL]

If you follow them, your ears will thank you. The two of those raise the cost to $977.03, plus approximately $27.10 shipping and handling, but you get a $50 mail-in rebate from the PSU and given its energy efficiency and tight voltage regulation, which can be found in the following review, it is a steal for $34.99 after the $50 mail-in rebate:

[URL]http://www.silentpcreview.com/article263-page1.html[/URL]

In total, that should cost $977.03 + $27.10 - $50 + $0.42 = $954.55. You can bring that down by approximately 2% if you get cashback through fatwallet and an additional 1% to 2% percent if you use a special credit card:

[URL]http://www.fatwallet.com/cash-back-coupons/Newegg[/URL]
[URL]https://www.penfed.org/productsAndRates/creditCards/RewardCards.asp[/URL] - this one lowers your bill by 1.25%, it is the one I use

If you were to use the 1.25% cashback credit card and get cashback through fatwallet, the total cost will be floor(($977.03 + $27.10 - $50) * 0.98 * 0.9875 + 0.5) + $0.42 = $923.78.

That will give you all of the components you will need to build the system you specified. You might also want things like an optical drive, a floppy drive, a keyboard, a mouse, a monitor, speakers, a printer, etcetera, but you probably will be able to get those from an older system.

ShiningArcanine, the "CORSAIR ValueSelect 2GB (2 x 1GB) 240-Pin DDR2 SDRAM DDR2 667 (PC2 5300) Desktop Memory" memory you recommend is a no go for a quad core : it is way to slow. (I would even go as far as saying it is to slow for a core2 duo dual core.)

Do not forget the requirement : "- FAST! Processor and FAST! Memory"

With PC2 8500 memory you will get a 60% performance boost. It is more expensive of course. 89,00 USD for the "CORSAIR XMS2 2GB (2 x 1GB) 240-Pin DDR2 SDRAM DDR2 1066 (PC2 8500) Dual Channel Kit Desktop Memory Model TWIN2X2048-8500C5" and 114,00 USD for the "CORSAIR Dominator 2GB (2 x 1GB) 240-Pin DDR2 SDRAM DDR2 1066 (PC2 8500) Dual Channel Kit Desktop Memory Model TWIN2X2048-8500C5D".

And forget everything people are saying at anandtech and so on. They do not use the CPU's like you will. They test memory throughput by loading one instance of their test program... Their conclusion is of course that PC2 6400 memory is good enough.

Jacob

A couple comments I have read recently on this forum suggest:

1. Wait for 45nm chips. I understand the Q6600 is still 65nm

2. AMD Quads are closer to 4 times than the Intel Quads.

Truth or Rumor?

[QUOTE=petrw1;120525]A couple comments I have read recently on this forum suggest:

1. Wait for 45nm chips. I understand the Q6600 is still 65nm

2. AMD Quads are closer to 4 times than the Intel Quads.

Truth or Rumor?[/QUOTE]The 45nm chips are already here... But at which price ! 1100,00 USD or so.

At the moment AMD cannot deliver their quadcores because of a production bug with the Level 3 cache. I also read some reviews (anandtech I believe) that state that the AMD is not performing as it was hoped (and much slower than Intel products.) Of course there is still some type of work where AMD shines (factoring ans especially with 64 bit programs I believe.)

Jacob

The complete line of Intel 45nm (Penryn) cpus are on the verge of being offered to the public. The Q9450 (quad-core) looks like the best bang for the buck - 2.66Ghz with 12Mb cache for $316.
[URL]http://www.overclock.net/intel-cpus/266534-best-45nm-price-performance.html[/URL]


Match it with this memory for $47 (after mail-in rebate), and you will have one fast pc.
[URL]http://www.newegg.com/Product/Product.aspx?Item=N82E16820145034[/URL]


The Q6600 will be going away soon.

A mild overclock on the Q9450 would get it up to the same frequency as the QX9650 that costs $1,000.

[QUOTE=ShiningArcanine;120515]high quality integrated graphics [/QUOTE] Just don't go with integrated GFX as it will occupy some of the memory bandwidth needed by CPU...
Also AFAIR there were some general perfomance issues with integrated chipsets and Prime95.

[quote=petrw1;120474]I will probably buy it in the spring.
Advice???[/quote] Wait until spring, or when you are ready to buy and re-ask the question as many things will change between now and then. ;)

Personally I think there are not too many software well support vista yet,&it's funny to say it as a memory pig:lol:

[quote]…it's funny to say it as a memory pig:lol:[/quote]
In most cases it appears that way because it caches all free memory, like it should. (Why waste "free" memory?)

We've seen it go as low as 2-3MB free, and we don't run any heavy applications. We also have a 4GB ReadyBoost cache as well.

We sure wish we had the 64-bit version. The 32-bit version doesn't see all of our memory.

:down:

[QUOTE=S485122;120528]The 45nm chips are already here... But at which price ! 1100,00 USD or so.

At the moment AMD cannot deliver their quadcores because of a production bug with the Level 3 cache. I also read some reviews (anandtech I believe) that state that the AMD is not performing as it was hoped (and much slower than Intel products.) Of course there is still some type of work where AMD shines (factoring ans especially with 64 bit programs I believe.)[/QUOTE]

Nice article here about AMD’s recent woes and near-term prospects here:

[url=http://bigtech.blogs.fortune.cnn.com/2007/12/12/has-intel-crushed-amd]Fortune Big Tech: Has Intel Crushed AMD?[/url]

[quote=petrw1]2. AMD Quads are closer to 4 times than the Intel Quads.

Truth or Rumor?[/quote]

Since AMD isn't shipping Barcelona in volume yet, it would have to be classified as "rumor". Also, based on some hot-of-the-presses numbers I just got from my buddies at Sun for my in-dev multithreaded Mlucas code running on 4 and 8-core Intel servers, Intel's quad-core solution looks pretty darn good. And note that I do hope AMD can fix the bugs plaguing Barcelona soon. Healthy competition - it's, well, healthy. But at least in the near-term, Intel's decision to leverage their proven dual-core into a quickie quad and keep doing what they are geniuses at - shrinking the process - appears to have been the right call. Might be a nice opportunity to pick up some AMD stock on the cheap, though. ;)

[QUOTE=petrw1;120474]I will probably buy it in the spring.

Requirements:
- FAST! Processor and FAST! Memory
- Probably at least a quad (or are 8's out and affordable now?)
- Prime use will be intensive computing: i.e. GIMPS and related stuff.
Advice???[/QUOTE]

- as pointed out, quads suffer from a memory bottleneck.
- fast memory lanes and fast processors, really a big increase in speed will be in 2009 with the introduction of a new Intel processor design. Like the Core 2 was a big improvement over the flawed Pentium IV design.
- So you might want to spend your money in 2009 and buy a cheap model in 2008, which you then give to your little brother a year later.

The Intel NEHALEM architecture will debut in 2009, with an integrated memory controller, including an octo-core cpu.

The memory bottleneck will be somewhat relieved. Maybe the cache sizes will increase to the point where a memory bottleneck is no longer relevant to PRIME95.

[QUOTE=S485122;120518]ShiningArcanine, the "CORSAIR ValueSelect 2GB (2 x 1GB) 240-Pin DDR2 SDRAM DDR2 667 (PC2 5300) Desktop Memory" memory you recommend is a no go for a quad core : it is way to slow. (I would even go as far as saying it is to slow for a core2 duo dual core.)

Do not forget the requirement : "- FAST! Processor and FAST! Memory"

With PC2 8500 memory you will get a 60% performance boost. It is more expensive of course. 89,00 USD for the "CORSAIR XMS2 2GB (2 x 1GB) 240-Pin DDR2 SDRAM DDR2 1066 (PC2 8500) Dual Channel Kit Desktop Memory Model TWIN2X2048-8500C5" and 114,00 USD for the "CORSAIR Dominator 2GB (2 x 1GB) 240-Pin DDR2 SDRAM DDR2 1066 (PC2 8500) Dual Channel Kit Desktop Memory Model TWIN2X2048-8500C5D".

And forget everything people are saying at anandtech and so on. They do not use the CPU's like you will. They test memory throughput by loading one instance of their test program... Their conclusion is of course that PC2 6400 memory is good enough.

Jacob[/QUOTE]

As far as lucas lehmer tests go, there will be no memory fast enough until there is a computer processor with n * 8MB of L2 cache where n is its number of cores.

Then that will be fast enough until the numbers become so big that n instances of Prime95 exceed the storage capabilities of the n * 8MB of L2 cache, at which point we will need n * 16MB of L2 cache.

[QUOTE=Cruelty;120548]Just don't go with integrated GFX as it will occupy some of the memory bandwidth needed by CPU...
Also AFAIR there were some general perfomance issues with integrated chipsets and Prime95.[/QUOTE]

The integrated graphics in Intel's chipsets have had hardware rendering since the G965 chipset. The motherboard I suggested has the latest integrated graphics chipset from Intel. He could play games like Far Cry on it if he wants and unless intel has released new drivers that offload additional processing to the CPU to raise the frame rates, all of the work is done by a dedicated graphics core.

[url]http://blogs.intel.com/technology/2007/08/gaming_on_integrated_graphics.php[/url]

[QUOTE=IronBits;120575]Wait until spring, or when you are ready to buy and re-ask the question as many things will change between now and then. ;)[/QUOTE]

That would be the best thing to do.

The primary issue with Prime95 and integrated graphics is the use of main memory for the framebuffer; it can consume a [formerly huge (think PC100 days), now less substantial] portion of the memory bandwidth. The additional CPU cycles needed to render the images remains minor in comparison.

[QUOTE=ShiningArcanine;120634]As far as lucas lehmer tests go, there will be no memory fast enough until there is a computer processor with n * 8MB of L2 cache where n is its number of cores.[/QUOTE]I agree.

But considering that :
-with quad cores the performance for LL is proportional to the memory speed (tested with 5400, 6400, 8500 memory),
- the original query was about "FAST!" memory,
the proposals of 5300 or 6400 memory did not meet the request.

8500 memory is not the fastest currently available but it is still affordable, if price was not a problem the answer to the query would have been a DDR3 board with DDR3 1800 (PC3 14400) memory but that kind of memory is priced in the 500,00 USD range.

Jacob

[QUOTE=S485122;120644]8500 memory is not the fastest currently available but it is still affordable, if price was not a problem the answer to the query would have been a DDR3 board with DDR3 1800 (PC3 14400) memory but that kind of memory is priced in the 500,00 USD range.[/QUOTE]

Is there any point in having a memory faster than the FSB? I think low latency might be better than high bandwidth for P95. Anybody out there with some concrete numbers?

[QUOTE=sdbardwick;120641]The primary issue with Prime95 and integrated graphics is the use of main memory for the framebuffer; it can consume a [formerly huge (think PC100 days), now less substantial] portion of the memory bandwidth. The additional CPU cycles needed to render the images remains minor in comparison.[/QUOTE]

The FSB and memory bus speeds run asyncronously on recent Intel motherboards, so the memory bandwidth requirements of the integrated graphics core should be negligible for 2D applications, unless he overclocks the FSB to match the memory bus.

[QUOTE=S485122;120644]I agree.

But considering that :
-with quad cores the performance for LL is proportional to the memory speed (tested with 5400, 6400, 8500 memory),
- the original query was about "FAST!" memory,
the proposals of 5300 or 6400 memory did not meet the request.

8500 memory is not the fastest currently available but it is still affordable, if price was not a problem the answer to the query would have been a DDR3 board with DDR3 1800 (PC3 14400) memory but that kind of memory is priced in the 500,00 USD range.

Jacob[/QUOTE]

In that case, he should get this motherboard:

[url]http://www.asus.com/products.aspx?l1=3&l2=11&l3=534&l4=0&model=1729&modelmenu=1[/url]


It has the lowest latencies avaliable by nature of having the ICs soldered onto the board, although avaliability will be a problem and he will need a discrete graphics accelerator as Asus does not design its products to cater to the scientific community.

[QUOTE=axn1;120662]Is there any point in having a memory faster than the FSB? I think low latency might be better than high bandwidth for P95. Anybody out there with some concrete numbers?[/QUOTE]

All of the electrons in a computer travels at a fixed rate, roughly 0.9c. Computers use pulses, that operate at certain frequencies to send data over fixed pathways. Each cycle takes up a certain amount of time and with each cycle, a certain amount of data is sent over these pathways. If you increase the frequencies at which the computer operates, say by a factor of two, the amount of data transfered per cycle will not change, but the number of cycles per given time unit will change and since the number of cycles per given time unit in this example have increased by a factor of two, the time it takes from the moment the first unit of data is received and the last unit of data is received, from a given transmission, is cut by a factor of two. The time it takes between sending the first unit of data and receiving the first unit of data, will however, remain constant. Because of this, latencies go down with higher memory transfer rates, but there is a limit to how much it can go down.

[QUOTE=ShiningArcanine;120664]All of the electrons in a computer travels at a fixed rate, roughly 0.9c.[/QUOTE]

Ah, no - actually it`s the [b]electric field[/b], i.e. the voltage-differential-induced electromagnetic [url=http://google.com/answers/threadview?id=345010]current wave[/url], which propagates at the speed of light [whose value in conducting solids is typically ~2/3 that of its vacuum value] - IIRC the actual *electrons* move [on average] exceedingly slowly - some discussion of that [url=http://amasci.com/miscon/speed.html]here[/url]. To use a macroscopic analogy, a tsunami moves across the ocean at hundreds of km/hour, even though the actual water molecules never move anywhere near that fast, and for the most part [until the wave breaks] simply do a circular motion as the wave passes and wind up where they started. [This is also what happens for AC power, which is why AC rather than DC is used for large-scale power distribution: Tesla understood this, Edison failed to]. In such cases it`s *energy* that propagates at high speed, not particles. [Admittedly, "particle" can't be divorced from "wave" for electrons, but we can still discuss the maximum amplitude of the QM wavefunction as if it were a point particle of mass.]

[QUOTE=ewmayer;120671]Ah, no - actually it`s the [b]electric field[/b], i.e. the voltage-differential-induced electromagnetic [url=http://google.com/answers/threadview?id=345010]current wave[/url], which propagates at the speed of light [whose value in conducting solids is typically ~2/3 that of its vacuum value] - IIRC the actual *electrons* move [on average] exceedingly slowly - some discussion of that [url=http://amasci.com/miscon/speed.html]here[/url]. To use a macroscopic analogy, a tsunami moves across the ocean at hundreds of km/hour, even though the actual water molecules never move anywhere near that fast, and for the most part [until the wave breaks] simply do a circular motion as the wave passes and wind up where they started. [This is also what happens for AC power, which is why AC rather than DC is used for large-scale power distribution: Tesla understood this, Edison failed to]. In such cases it`s *energy* that propagates at high speed, not particles. [Admittedly, "particle" can't be divorced from "wave" for electrons, but we can still discuss the maximum amplitude of the QM wavefunction as if it were a point particle of mass.][/QUOTE]

Computer processors are DC circuits. I am not an electrical engineering major, but according to the following web page, transistors "cannot switch AC:"

[url]http://www.kpsec.freeuk.com/trancirc.htm[/url]

Also, in a DC circuit, electrons are always moving in a current with their average position being the integral of their average velocity while in an AC circuit, their position is a sine function involving time, a frequency and a phase angle, mutipled by some amplitude and their velocity at Pi minus the phase angle plus any mutiple of two Pi being zero. In an AC circuit, talking about currents is meaningless because each electron never leaves a sphere with a radius that is the amplitude of the wave function and an origin that is the point where its velocity is zero.

Despite that, I was wrong. Not all of the electrons in a computer travel at a fixed rate (which is actually an average rate), as some of the electrons in the PSU travel back and forth, with no net change in displacement worth mentioning.

By the way, that is interesting, but it does not answer axn1's question. Getting back to his question, I would like to add some information I omitted:

Low timing memory reduces the amount of time from the memory receiving a request to the memory sending a request, but with DDR2 memory (and I believe even more profoundly with DDR3 memory), many times lower clock rates hurt latencies more than lower timings help.

Also, "All of the electrons in a computer travels at a fixed rate, roughly 0.9c" should have been "All of the electrons in a computer travel at a fixed rate, roughly 0.9c." I apologize for my misuse of the English language.

[quote=ShiningArcanine;120685]Computer processors are DC circuits.

[/quote]

They most certainly are not!

Transistors in a CPU must talk to each other, and they do so over tiny wires laid out in the silicon. The bits transmitted back and forth are approximations of square waves, limited by the bandwidth of the transistor. These are AC signals.

[quote=ShiningArcanine;120685]
transistors "cannot switch AC:"
[/quote]

I don't know what you're talking about here. If you stick an AC signal into a transistor's gate (or base, if you're thinking of BJT's for some reason), you'll get a square wave out if the AC signal swings through the appropriate threshold voltage.


[quote=ShiningArcanine;120685]
Also, "All of the electrons in a computer travels at a fixed rate, roughly 0.9c" should have been "All of the electrons in a computer travel at a fixed rate, roughly 0.9c." I apologize for my misuse of the English language.[/quote]

Again, when you transmit a bunch of bits from A to B, the electrons in the transmission medium barely move. The field's propagate. In motherboards, the speed is closer to 1/2c, unless they are being designed with fairly exotic dielectric materials (read: expensive).

The holes move. Actual movement of indvidual electrons is less important than the speed of the holes.

[url]http://en.wikipedia.org/wiki/Electron_hole[/url]

[quote=ewmayer;120671]...simply do a circular motion as the wave passes and wind up where they started. [This is also what happens for AC power, which is why AC rather than DC is used for large-scale power distribution: Tesla understood this, Edison failed to]. [/quote]
Don't transformers enter the explanation of why AC is used in a
crucial way? Or is your explanation related to this in a way I have
yet to fathom? Power dissipated in the cables is I[sup]2[/sup]R whether it is
AC or DC.

[QUOTE=bsquared;120692]They most certainly are not!

Transistors in a CPU must talk to each other, and they do so over tiny wires laid out in the silicon. The bits transmitted back and forth are approximations of square waves, limited by the bandwidth of the transistor. These are AC signals.[/QUOTE]

I thought that they used voltage thresholds where anything below a value k would be 0 and anything equal to or greater than that value would be 1 and that motherboards have DC-DC circuitry to step down the voltage of the current flowing into a processor to the voltage necessary for the processor to operate. I thought all of this meant that a DC current was running through the processor.

[QUOTE=bsquared;120692]I don't know what you're talking about here. If you stick an AC signal into a transistor's gate (or base, if you're thinking of BJT's for some reason), you'll get a square wave out if the AC signal swings through the appropriate threshold voltage. [/QUOTE]

If you use an AC signal, can you get the transistor to operate as it is does with a DC signal?

[QUOTE=bsquared;120692]Again, when you transmit a bunch of bits from A to B, the electrons in the transmission medium barely move. The field's propagate. In motherboards, the speed is closer to 1/2c, unless they are being designed with fairly exotic dielectric materials (read: expensive).[/QUOTE]

I was probably thinking of beta radiation when I said 0.9c. I did not think there was a difference between the speed of beta radiation and the speed of electrons in a wire.

[quote=ShiningArcanine;120729]I thought that they used voltage thresholds where anything below a value k would be 0 and anything equal to or greater than that value would be 1 and that motherboards have DC-DC circuitry to step down the voltage of the current flowing into a processor to the voltage necessary for the processor to operate. I thought all of this meant that a DC current was running through the processor.
[/quote]

[U][COLOR=#22229c][URL]http://en.wikipedia.org/wiki/CMOS[/URL][/COLOR][/U]
Transistors have threshold voltages, yes. As referenced above, CMOS is the processing technology used in microprocessors. Roughly, a bunch of transistors are connected to positive and negative rails (Vdd and Vss), and as the gate voltages are swung through the transistion voltage, the output swings from one rail to the other. The rails are DC voltages. Each state, 0 or 1, is a DC voltage. But the transistion between states creates an AC signal. There is a transient current pulse with every bit transistion which makes for a very non-DC current profile. This is why your motherboard has hundreds of tiny surface mount capacitors surrounding the CPU. They are there to help alleviate the enormous transient current demand of the CPU due to millions of transistors changing state nearly simultaneously.

[quote=ShiningArcanine;120729]
If you use an AC signal, can you get the transistor to operate as it is does with a DC signal?
[/quote]

Sure, there is nothing different. A DC signal is just a slow AC one ;) Seriously though, if you are doing an experiment with a transistor where you dial up the gate voltage with a lab supply, for instance, and watch for a change in output current; that is also an AC signal. Just a very slowly varying one. As the signals get faster and faster, you just need better and better transistors to be able to see the transitions.


[quote=ShiningArcanine;120729]
I was probably thinking of beta radiation when I said 0.9c. I did not think there was a difference between the speed of beta radiation and the speed of electrons in a wire.[/quote]

The speed of the *field* in the wire (not the electrons) is equal to 1/sqrt(epsilon_r * mu_r), where epsilion_r and mu_r are the relative permittivity and permeability of the dielectric material surrounding the transmission line. For basic motherboards, this is usually FR-4 or some variant, which almost always has epsilion_r of about 4. Thus the speed is about 1/sqrt(4) = 1/2 slower than light in a vacuum.

[quote=davieddy;120723]Don't transformers enter the explanation of why AC is used in a
crucial way? Or is your explanation related to this in a way I have
yet to fathom? Power dissipated in the cables is I[sup]2[/sup]R whether it is
AC or DC.[/quote]

[URL]http://en.wikipedia.org/wiki/Power_distribution[/URL]

Power is also V[sup]2[/sup]/R.

Transformers are there to increase the voltage at the expense of the current. The net power stays the same, but you don't want large amounts of current sloshing around in the wires because large currents create large magnetic fields and also tend to melt copper. The use of AC is what fundamentally allows the use of transformers.

[quote=bsquared;120735][URL]http://en.wikipedia.org/wiki/Power_distribution[/URL]

Power is also V[sup]2[/sup]/R.

[/quote]


Power dissipated in the cables is only V[sup]2[/sup]/R
if you interpret "V" as the potential drop along the cable
V=I*(resistance of cable).

It's closer to summer than spring ... asking again
 

[QUOTE=IronBits;120575]Wait until spring, or when you are ready to buy and re-ask the question as many things will change between now and then. ;)[/QUOTE]

Anything big and new and relevant to mention since 4 months ago?

[quote=axn1;120477]Processor of choice: [URL]http://www.tigerdirect.ca/applications/SearchTools/item-details.asp?EdpNo=2643933&Sku=CP2-DUO-Q6600[/URL][/quote]

Q6700 was a ridiculous follow-up to Q6600, but [B]Q9300[/B] looks sweet (value for money). Buy it bare OEM and get a Tuniq Tower cooler or some other nice cooler -
[URL]http://www.newegg.com/Product/Product.aspx?Item=N82E16819115040[/URL]
[URL]http://www.newegg.com/Product/Product.aspx?Item=N82E16835154001[/URL]
...and get it to run at 3.2-3.6GHz.

I disagree that C2Q's "perform not much better than C2D". They scale decently to 4 threads. {{Rant}} I wish I could say the same about my work comp which is an dual-quad-Xeon: that one scales horribly. Still, I run it on 7 threads (7 different LLs! not 7 threads on 1 LL - that's a complete waste) which is marginally better than 6 or 5. Worse than 8, though. {{/rant}}

At home I run a Q6600 with Corsairs PC2-8500s and all what people said above is true -- you cannot afford to go lower than 8500 memory these days. But I have a complicated relationship with Corsairs - they keep switching numbers and voltages (while selling it as the same part); I bought 2Gb more and they don't quite match -same part number, different voltages! (I need to play some more with BIOS). Maybe you could do better with a G.SKILL or Kingston.

Well, my 2 cents.

[QUOTE=petrw1;131940]Anything big and new and relevant to mention since 4 months ago?[/QUOTE]I'd recommend an Intel 9000-series (45nm) quadcore for the CPU (e.g. [url=http://www.canadacomputers.com/index.php?do=ShowProduct&cmd=pd&pid=018244&cid=CPU.84]Q9300[/url] or [url=http://www.canadacomputers.com/index.php?do=ShowProduct&cmd=pd&pid=018295&cid=CPU.84]Q9450[/url]). The Q9450 is likely worth the extra ~$60, partly for the extra 166MHz clock speed, but mostly for the doubling of cache from 6MB to 12MB.

As for RAM, you want it as fast as you can afford, at least DDR2-1066. Some nice fast DDR3 would be nice, but it's still pretty expensive compared to DDR2, although prices have fallen significantly in the last few months ([url=http://www.canadacomputers.com/index.php?do=ShowProduct&cmd=pd&pid=017746&cid=RAM.346.754]4GB of DDR2-1066[/url] is still a fair bit cheaper than [url=http://www.canadacomputers.com/index.php?do=ShowProduct&cmd=pd&pid=016973&cid=RAM.346.539]2GB of DDR3-1600[/url]).

You can hedge your bets with a board that supports both DDR2 and DDR3 (e.g. [url=http://www.canadacomputers.com/index.php?do=ShowProduct&cmd=pd&pid=017991&cid=MB.157]Gigabyte GA-EP35C-DS3R[/url]) to use DDR2 now with the option to upgrade to DDR3 when prices drop.

Dell XPS
 

I am looking at a Dell XPS720.
2Q6600 Quad-Core
8MB L2 cache
2.4GHHz
1066FSB
3GB Dual channel DDR2 SDRAM at 800MHz-4 DIMMs

If I am testing in the 42,000,000 range on all four cores, can I expect to complete 4 numbers per month?

[url]http://www.mersenne.org/bench.htm[/url]

Strictly based on the statistics on the above link each core could theoretically finish 42,000,000 in 30 days. IF it can run 24 hours a day with nothing much else taking CPU time.

However, based on previous posts in this forum and other related hardware/benchmark forums you will NOT get 100% throughput on each core if all 4 are running primenet.

How much would performance increase with a Q9450 and DDR3 memory?
Any idea in cost increase?

[QUOTE=petrw1;132124]However, based on previous posts in this forum and other related hardware/benchmark forums you will NOT get 100% throughput on each core if all 4 are running primenet.[/QUOTE]For a Core2Quad with DDR2-800 you can expect an overall throughput with 4 cores of somewhere around 2.8x throughput of a single core. Based on some [url=http://www.mersenneforum.org/showthread.php?t=10208]rough benchmarks[/url] I took yesterday, a Q9x50 series with fast DDR3 could be ~20% faster with 3 or 4 threads than a Q6600, so you should get at least 3 exponents processed per month, maybe a little more.

[quote=James Heinrich;132140]For a Core2Quad with DDR2-800 you can expect an overall throughput with 4 cores of somewhere around [B]2.8x[/B] throughput of a single core.[/quote]
Timed my Q6600 (@3200MHz) with [I]DDR2-1066[/I] (set FSB multiplier to 3:2):
For the popular 2048M FFT range (under 39.5M) range for a point of reference -
1 LL thread - 0.036s/iter
2 LL threads scale to 1.95x of a single core (0.037s/iter if you get the right pair of cores)
3 LL threads scale to 2.66x of a single core (0.042s/iter on a pair and 0.037s/iter on a one = 0.0405s/iter harmonic mean)
4 LL threads scale to [B]3.16x[/B] of a single core (0.0455s/iter)

The dual quad core at work scales much worse: all 8 threads (8 LLs) produce less work than 7 LLs. Sad, but true. So I run 7 LLs. (but this, combined, is still better than 6, 5, ..., 1) You have to use harmonic means of iteration timings reported in threads (obviously).

DDR2-1066 don't break the bank. Do DDR3 if you are rich.

Please post your detailed system components list. It would be helpful if you would also post your overclocking settings in BIOS.

I'm interested in adding to my P95 farm, and the iteration times when running four LL tests looks quite good.

[quote=rx7350;132183]Please post your detailed system components list. It would be helpful if you would also post your overclocking settings in BIOS.

I'm interested in adding to my P95 farm, and the iteration times when running four LL tests looks quite good.[/quote]

Sure, too easy (as the Australians say) - [URL]http://valid.x86-secret.com/show_oc.php?id=339536[/URL]

I believed in these principles -
1) weird multiplers (like 8:5 or 5:3) are not really working for memory, and
2) "thou shall not touch your memory's frequency" (the vendor OC'd it for you already! :grin: Especially if their name is Corsair)
=> so I had only two or three FSB frequencies to test.
266 x 4.000 = 1066MHz (original) 266 x 9 = 2400MHz (CPU)
320 x 3.333 = 1066MHz
333 x 3.200 = 1066MHz
355 x 3.000 = 1065Mhz

I liked the last one the best. The above timings are from this configuration.
In BIOS you need to know the Ctrl-F1 combination (it's not secret, it's in TFM) - and see the attached snapshot.

P.S. Memory shows in BIOS as DDR2-800 but it is sold by Corsair as XMS-1066
The parts list (built in December) -
[URL]http://www.newegg.com/Product/Product.aspx?Item=N82E16819115017[/URL]
[URL]http://www.newegg.com/Product/Product.aspx?Item=N82E16813128059[/URL]
[URL]http://www.newegg.com/Product/Product.aspx?Item=N82E16811129021[/URL]
[URL]http://www.newegg.com/Product/Product.aspx?Item=N82E16817153039[/URL]
[URL]http://www.newegg.com/Product/Product.aspx?Item=N82E16820145043[/URL] and an HD and a video card

Added in March after getting addicted to GIMPS -
[URL]http://www.newegg.com/product/product.aspx?Item=N82E16835154001[/URL] (very important, without it I had Tj=[B]70C+[/B] under the 4 LLs @2400MHz; now I have 60C under the 4 LLs @3200MHz) there are other good coolers too. But this one is the king.

P.P.S. And now is not December - it's time to buy [B]Q9450[/B] (or Q9300 at least) instead of Q6600.
Maybe there's a DDR2-1111 or 1200 choice already, as well. [URL="http://www.mersenneforum.org/showpost.php?p=132038&postcount=32"]Up a notch everything[/URL], of course.

what would be really cool is if prime95 let you use readyboost as far as I know you can't if I could I'd be 5 times as fast lol.