What I have Learned about Computer Performance from SPCR

The forum for non-component-related silent pc discussions.

Moderators: NeilBlanchard, Ralf Hutter, sthayashi, Lawrence Lee

ces
Posts: 3395
Joined: Thu Feb 04, 2010 6:06 pm
Location: US

What I have Learned about Computer Performance from SPCR

Post by ces » Tue Feb 08, 2011 3:41 am

I though I would compile a summary of what I have learned about computer performance by asking questions at SPCR.

RULES OF THUMB
alecmg wrote:My rule of thumb is

1. Cache is good for gaming

2. Number of cores is good for encoding

3. GHz is good for responsiveness.

4. memory frequency/timings are important in all three

5. Memory amount only matters when its not enough.
NICE SUMMARY (though doesn't cache also help encoding?)

For more see:
More Cache - What good is it?
viewtopic.php?f=28&t=58469

ces
Posts: 3395
Joined: Thu Feb 04, 2010 6:06 pm
Location: US

Re: What I have Learned about Computer Performance from SPCR

Post by ces » Tue Feb 08, 2011 3:50 am

ces wrote:From Benchmark Reviews: "Taking a small collection of CPU coolers, Benchmark Reviews tested the performance of four different fans.":
1. Scythe Kaze-Jyuni Slip-Stream model SY1225SL12SH (110.31 CFM Advertised @ 37.0 dBA) 12V/0.53A 120x120x25mm 1900 RPM
2. CoolAge SX2 fan, model CA-120SX2 (80 CFM Advertised @ 32 dBA) 12V/0.19A 120x120x38mm
3. Yate Loon model D12SH-12 (88 CFM @ 40 dBA) 12V/0.30A 120x120x25mm 2200 RPM
4. Scythe Ultra Kaze model DFS123812H-3000 (133.6 CFM @ 45.9 dBA) 12V/0.60A 120x120x38mm 3000 RPM
(results reported in that order)

Thermal Difference (°C over ambient)
Prolimatech Megahalems 38.0°C; 39.1°C; 36.3°C; 36.2°C;
Scythe Mugen-2 SCMG-2000 39.9°C; 41.0°C; 38.6°C; 38.8°C;
Xigmatek Thor's Hammer S126384 40.7°C; 43.5°C; 39.1°C; 41.6°C;
Titan FENRIR TTC-NK85TZ 43.6°C; 45.0°C; 42.5°C; 43.2°C;
CoolAge FO-X120TF Wind Tunnel 44.6°C; 44.2°C; 42.6°C; 43.3°C;
Thermolab BARAM 45.5°C; 46.2°C; 42.8°C; 43.4°C;

http://benchmarkreviews.com/index.php?o ... mitstart=5
The Yate Loon model D12SH-12 (88 CFM @ 40 dBA) 12V/0.30A 120x120x25mm 2200 RPM was Benchmark's the best performing fan. That is interesting as this is basically SPCR's Nexus reference fan operating at 2200 rpm as opposed to its standard 1000 rpm.

It seems that for the most cooling per dB, whether on a CPU cooler requiring good static pressure performance, or for a case fan, it is pretty hard to go wrong with the SPCR reference Nexus fan.

alecmg
Posts: 204
Joined: Thu Mar 13, 2008 5:56 am
Location: Estonia

Re: What I have Learned about Computer Performance from SPCR

Post by alecmg » Tue Feb 08, 2011 3:55 am

why thank you, not often do I get cited :)
Its a quick and dirty summary, can't be right in all cases.

Btw Sandy Bridge processors break these rules in many aspects.

ces
Posts: 3395
Joined: Thu Feb 04, 2010 6:06 pm
Location: US

Re: What I have Learned about Computer Performance from SPCR

Post by ces » Tue Feb 08, 2011 4:00 am

ces wrote:Optimizing Performance - Fast memory?

1. What hardware governs how quickly I can flit from window to window. Does the speed of the memory affect that?

2. What attributes of performance does fast memory help? What does it not help?
Answer to Question 1
swivelguy2 wrote:Performance flipping between windows is probably most dependent on:
1. Your monitor
2. Your mouse
3. Your GPU (2D performance)
Answer to Question 2
psyopper wrote:Other than a 1-2 % difference in most normal, non-synthetic tasks - not much. As pointed out, it's really handy for overclockers.
For more info see:
Faster Memory - What good is it?
viewtopic.php?f=28&t=58467
Last edited by ces on Tue Feb 08, 2011 4:20 am, edited 1 time in total.

ces
Posts: 3395
Joined: Thu Feb 04, 2010 6:06 pm
Location: US

Re: What I have Learned about Computer Performance from SPCR

Post by ces » Tue Feb 08, 2011 4:07 am

Areas Critical to CPU Heatsink Performance:

From MikeC:
ces wrote:
MikeC wrote:
Big Pimp Daddy wrote:As a side benefit, it takes a lot longer for water to heat up than air, meaning that short periods of high processor activity will not raise system temperatures greatly, hence no need to increase fanspeeds for a while. In my aircooled system sporadic high loads will cause the CPU fan (which is temperature controlled) to speed up and down as the load changes. This is a lot more noticeable and intrusive than a constant speed.
This is simply a matter of increasing the hysteresis of the feedback loop... or making the fan run at a steady speed just high enough to keep the CPU in this side of throttling at full load -- but then it will be way cool enough at all other times and probably still pretty quiet if you have a big heatsink.
MikeC, given the performance of today's supercoolers, where do you think the gating factor to best cooling performance is? Is it:
(a) getting the heat from the silicon to the metal surface of the CPU unit,
(b) getting the heat from that surface to the base plate of the cooler (whether water or air based),
(c) getting the heat from the surface of the base plate to the cooling medium (whether heat pipe or water),
(d) getting the heat from the cooling medium (whether heat pipe or water) to the fin array, or
(e) getting the heat from the fin array to the air?

It seems to me, just subjectively touching various parts of heat sinks, that contrary to my visual intuition, the fin arrays may be less critical than other, more subtle, parts of the heat sinks.
Hard to say, surely depends on the particular heatsink model. But not, generally speaking, a, b, or e. Considering the many hs that have come through, I think that c and d could be affected by the quality of soldering/fit. Variations in the fin array also have significant impact but are usually tied to total area and impedance.

For more info see:
Water Cooling Heresy
viewtopic.php?f=21&t=58028

ces
Posts: 3395
Joined: Thu Feb 04, 2010 6:06 pm
Location: US

Re: What I have Learned about Computer Performance from SPCR

Post by ces » Tue Feb 08, 2011 4:13 am

Jay_S wrote:
ces wrote:There is a problem with all the benchmarks out there. They don't appear to translate well to what an average user uses. They seem to be built around the curiosity of: (a) gamers, (b) high definition enthusiasts, and (c) hardware enthusiasts interested in synthetic benchmarks to chase.
Our hypothetical average user has no idea what cache is. The average user confuses memory and storage, thinks Windows comes with Office, and thinks POP is a beverage. They are not the intended audience for tech sites.

I agree - synthetics do not translate into real-world performance. Most tech sites use canned benchmarks because it's a convenient way to compare hardware. But others don't - for example, [H]ardOCP eschews canned benchmarks and instead actually plays games to test hardware. They report on the gaming experience. They're a PC gaming enthusiast site, and their tests translate perfectly into what their intended audience does.
ces wrote:Here is what I use. Typically with a fair number of windows open at the same time:
Browsers (Chrome and Foxfire and sometimes MS internet explorer)
Windows Explorer (I use it a lot and it sometimes has pauses while it waits for my HDD to be doing something - a black edition 1T WD)
Word processing (MS Office & Open Office - both can sometimes be slow in launching a file through Windows explorer)
Scanning software (bloated third party, bloated Brother, and extremely bloated HP software)
Anti-Virus (AVG and sometimes Avira - I have switched portions of them off because they can drain system resources)
X1 desktop search engine (when it is indexing files it loads the system up similarly to the Anti-virus software)
iDrive backup software (when it is backing up the system it loads the system similarly to the Anti-virus software)
PDF (Foxit PDF reader, Adobe PDF reader and CoolPDF printer)
farstone cloning/backup software
I'd say you typify the average office PC user. Most of your slowdowns sound like I/O issues, and will not be helped by larger CPU cache. The benchmarks related to cache were picked because they are CPU-intensive. Ie, the disks are waiting for the CPU to complete whatever it's doing. I don't think any of your aps behave this way.

I would get rid of AVG. AVG used to be the light-weight AV of choice, but has become bloated and slow over the past few years. I dropped AVG as soon as they started with the advert pop-ups (8? 8.5?). I use MS Security Essentials, which has been terrific to date. It's lightweight, fast, and well-reviewed.

Lots of windows, browsers and browser tabs will eat RAM. If you're using a 64-bit OS, don't be afraid to go a little nuts on RAM. If you're not overclocking, don't worry about memory clock speed. But pay attention to latencies. Get low-latency JDEC-compliant RAM, and lots of it.

The rest of your issues sound like
1) poor software coding
2) I/O limitations of your mechanical disks
ces wrote:I am building a new computer. I have a Zotac mini-ITX board. I have a Lian Li Q07 case. I was going to put a Q9550 CPU in it. But I want to power it with ElectroDacus's winmate PSU system. I am not fully comfortable using it to power the Q9550.

So I am looking for a less power hungry substitute 775 cpu. The published benchmarks do not seem to be helpful.
The three articles I linked to were a response to your question: "More cache, what good is it?", and is no help for the power question. The least power-hungy LGA775 CPU right now is probably one of the Wolfdale Celerons (xbit labs) precisely because they have less cache. I have an E3300 + G41 board in my HTPC, and it consumes no more power than the previous AMD 4050e (45W TDP) + 740G/SB700 system it replaced.
ces wrote:So I don't think this 775 system is going to be a speed upgrade to the Clarkdale system. But I don't want a sluggish system either.

So what CPU do I select? What is the tradeoffo in Cache and in GHz for my uses?
I don't think the LGA775 system will be faster than the Clarkdale system either. Most of your applications are not heavily threaded, so the Q9550's extra cores don't benefit you. Your aps look I/O bound. Faced with that I'd spend on a fast SSD and - to a lesser extent - high CPU clock speed for good single-threaded performance.

ces
Posts: 3395
Joined: Thu Feb 04, 2010 6:06 pm
Location: US

Re: What I have Learned about Computer Performance from SPCR

Post by ces » Tue Feb 08, 2011 4:21 am

alecmg wrote:Btw Sandy Bridge processors break these rules in many aspects.
Please explain further.

Mats
Posts: 3044
Joined: Fri Dec 26, 2003 6:54 am
Location: Sweden

Re: What I have Learned about Computer Performance from SPCR

Post by Mats » Tue Feb 08, 2011 4:26 am

A SSD seems like a quite important part, but it's left out in the list. Why?
I'd prefer getting a small SSD than having a 700 MHZ faster CPU.

ces
Posts: 3395
Joined: Thu Feb 04, 2010 6:06 pm
Location: US

Re: What I have Learned about Computer Performance from SPCR

Post by ces » Tue Feb 08, 2011 4:27 am

WATER COOLING - WHAT USE IS IT?

1. MOVING THE HEAT TO AMBIENT AIR

Whether you are direct air cooling or water cooling, don't you have to run the same thermal load into the ambient air whether through the air based fins or the water based fins to dispose of the heat into that air?

So then don't you need the same amount of air current to do so, at least on a sustainable basis?

And doesn't the water pump just add additional noise?

2. BENCHMARKED PERFORMANCE

Here look at the frostytech rankings
http://www.frostytech.com/articleview.c ... 501&page=5

The highest ranking water cooler is the CoolIT Domino ALC. It is out performed by a fair number of top line aircoolers, and it is loud! It is about 17db louder than the top ranked air cooler.

This is with a 150 watt thermal load.

I don't know how good or bad the CoolIT Domino ALC system is. It does perform well in the Frostytech rankings, but it is loud. And there are direct air systems that out perform it. And these direct air systems seem to be able to handle the heat load generated by any of todays CPUs, even overclocked ones.

3. COST

At NCIX:
the Noctua NH-D14 sells for $89.99 Canadian
the Sunbeam Tuniq Tower is 58.08 Canadian
the Zalman CNPS10X Extreme is $85.99 Canadian

4. AGAIN

Don't you have to move the same amount of air whether you are water cooling or using direct air?

Why would it be quieter moving this air through a water cooling radiator?

5. SAFETY

If your pump breaks what happens to your computer?

If one of the three fans on a Noctua NH-D14 breaks, what happens to your computer?

6. CAPACITY

Xbits has some data on the Noctua and the IFX 14
http://www.xbitlabs.com/articles/cooler ... d14_7.html

Their test platform is a challenging one:
Core i7-920, 4.01 GHz 1.36875

Their delta at full load using Linpack 64 is about 46C (70-24) using two 2000rpm fans. Delta at idle was about 10C (34-24). Probably you can get the same performance with less noise using 3 Apache PWM fans.

Three Apache PWM fans loping along at 500rpms under light to moderate load, probably generate almost no noise. They will kick in only under high stress conditions, and even then won't be that noisy. And light to moderate load probably includes gaming.

A Prolimatech with 2 Apaches will perform almost as well. As will a stock single fan Tuniq Tower 120 Extreme

Here is the performance data on stock single fan Tuniq Tower 120 Extreme performance at high and low speeds (using the single stock fan that comes with the heat sink):
http://www.frostytech.com/articleview.c ... 480&page=5

Note that at low speed it handles 150 watts just fine and is quieter than a standard Noctua Tower cooler.

7. WHAT AM I NOT GETTING? WHAT AM I MISSING?

I don't own a watter cooling setup. I admit I may be missing something. But what?

No one has yet articulated to me what I get in return for the greater complexity, cost, maintenance headaches and likelihood of failure?

And when a water cooling systems fails, it is likely to be a more catastrophic failure than losing one fan of two or three on a high end air cooler. Nothing good can come out of water leaking on a motherboard.

ces
Posts: 3395
Joined: Thu Feb 04, 2010 6:06 pm
Location: US

Re: What I have Learned about Computer Performance from SPCR

Post by ces » Tue Feb 08, 2011 4:29 am

Mats wrote:A SSD seems like a quite important part, but it's left out in the list. Why? I'd prefer getting a small SSD than having a 700 MHZ faster CPU.
What you say is more than valid. For most applications they are more important to performance than the CPU. The discussion was focused on CPUs, that is why SSDs were not brought up.
Last edited by ces on Tue Feb 08, 2011 6:17 am, edited 1 time in total.

Mats
Posts: 3044
Joined: Fri Dec 26, 2003 6:54 am
Location: Sweden

Re: What I have Learned about Computer Performance from SPCR

Post by Mats » Tue Feb 08, 2011 4:42 am

ces wrote:The discussion was focused on CPUs, that is why SSDs were not brought up.
That's what I thought at first too, then I saw #4 and #5, about RAM.
Xbitlabs compared adding a SSD to adding more RAM.

ame
Posts: 488
Joined: Tue Jul 10, 2007 10:35 pm
Location: Israel

Re: What I have Learned about Computer Performance from SPCR

Post by ame » Tue Feb 08, 2011 6:03 am

Mats wrote:
ces wrote:The discussion was focused on CPUs, that is why SSDs were not brought up.
That's what I thought at first too, then I saw #4 and #5, about RAM.
Xbitlabs compared adding a SSD to adding more RAM.

Sadly this review is tilted in a strange way. Since I know X-bit typically have good reviews it seems they just didn't ask the right question. This question is irrelevant for pro users. I understand the reasoning behind it as in "what should I buy - more RAM or an SSD" One could also answer - a better monitor. So really bad choice for a review topic IMO

More RAM is critical when you need it. The Photoshop test is the only test they did that needs the extra RAM. They should have come up with tests that use more RAM on Aftereffects, 3D and Video editing as well. But they didn't. I Guess its easy to find a megapixel image but not so easy to apply a stressful filter on some very high quality HD footage. Same with 3D apps, If they loaded a huge scene with huge textures....

If they had used 1GB VS 4GB RAM and an SSD the results were in favor of RAM across the board. Games would be unplayable. Windows would take forever to load. And video editing would be impossible...

SSD is great. But even though I own one, I can say its luxury rather than necessity.

ame
Posts: 488
Joined: Tue Jul 10, 2007 10:35 pm
Location: Israel

Re: What I have Learned about Computer Performance from SPCR

Post by ame » Tue Feb 08, 2011 6:10 am

CES on watercooling,

You've got a point. Heat has to go into the air eventually.

As a side note I've seen Intel, and a few extreme OC review sites, admit they haven't been able to get better OC from SNB using water cooling. Seems it has to do with the maximum clock possible regardless of cooling. Heat is probably not be the limiting factor anymore.

ces
Posts: 3395
Joined: Thu Feb 04, 2010 6:06 pm
Location: US

Re: What I have Learned about Computer Performance from SPCR

Post by ces » Tue Feb 08, 2011 6:12 am

alecmg wrote:My rule of thumb is...
5. Memory amount only matters when its not enough.[/i]
But it is a real killer when you need more memory and don't have it.

I guess to some extent, an SSD might moderate the hit... if the operating system does a decent job of swapping and substituting hard drive space for the missing memory.

ces
Posts: 3395
Joined: Thu Feb 04, 2010 6:06 pm
Location: US

Re: What I have Learned about Computer Performance from SPCR

Post by ces » Tue Feb 08, 2011 6:16 am

ame wrote:You've got a point. Heat has to go into the air eventually. As a side note I've seen Intel, and a few extreme OC review sites, admit they haven't been able to get better OC from SNB using water cooling. Seems it has to do with the maximum clock possible regardless of cooling. Heat is probably not be the limiting factor anymore.
Take a look at this review of the Corsair Hydro H70 Liquid CPU Cooler
http://benchmarkreviews.com/index.php?o ... mitstart=4

MikeC
Site Admin
Posts: 12285
Joined: Sun Aug 11, 2002 3:26 pm
Location: Vancouver, BC, Canada
Contact:

Re: What I have Learned about Computer Performance from SPCR

Post by MikeC » Tue Feb 08, 2011 8:55 am

ces wrote:The Yate Loon model D12SH-12 (88 CFM @ 40 dBA) 12V/0.30A 120x120x25mm 2200 RPM was Benchmark's the best performing fan. That is interesting as this is basically SPCR's Nexus reference fan operating at 2200 rpm as opposed to its standard 1000 rpm.
Actually, this is not true. There are enough differences between the standard YL 120 fan and the Nexus 120 to make them quite different animals. I believe those differences are well documented in our review of the Nexus.... way back when in oue of our first 120mm fan roundups.
It seems that for the most cooling per dB, whether on a CPU cooler requiring good static pressure performance, or for a case fan, it is pretty hard to go wrong with the SPCR reference Nexus fan.
Agreed.

fumino
Posts: 298
Joined: Thu Aug 05, 2010 4:38 pm
Location: ontario

Re: What I have Learned about Computer Performance from SPCR

Post by fumino » Tue Feb 08, 2011 2:04 pm

when it comes to cooling, unless you're prepared to spend at least $250 on proper water cooling, youre going to get better performance(per dollar/watt/dB) from air cooling.

ces
Posts: 3395
Joined: Thu Feb 04, 2010 6:06 pm
Location: US

Re: What I have Learned about Computer Performance from SPCR

Post by ces » Wed Feb 09, 2011 7:16 am

MikeC wrote:Actually, this is not true. There are enough differences between the standard YL 120 fan and the Nexus 120 to make them quite different animals. I believe those differences are well documented in our review of the Nexus.... way back when in oue of our first 120mm fan roundups.
My understanding is that the Yate Loons are all based on a shared reference design, with basically two types of blades (and probably some variations on that) and different electronics/electricals. Not unlike the family of Scythe Slipstreams and the family of Nidec Gentle Typhoon fans.

What is unigue to the the SPCR Nexus reference fans is that that have some unique electronics/electricals that moderate motor noise at low RPMs.

Is that correct?

It is my expectation that, subject to testing, it would not be unreasonable to expect the other positive characteristics of the SPCR Nexus reference fans to likely be present in other derivations of the Yate Loon reference design.

Do you have any opinion as to generally, how fans built on the Yate Loon reference design compare generally to fans built on the Scythe Slipstream reference design? It seems to me not unreasonable to expect slightly better static pressure performance at equal RPMs from the Yate loons, at least in those Yate Loons using the same fan blade design as the SPCR Nexus reference fans.

Also, do you know if there is any decernable difference between the red bladed Nexus fans and the white bladed ones?

MikeC
Site Admin
Posts: 12285
Joined: Sun Aug 11, 2002 3:26 pm
Location: Vancouver, BC, Canada
Contact:

Re: What I have Learned about Computer Performance from SPCR

Post by MikeC » Wed Feb 09, 2011 7:55 am

ces wrote:
MikeC wrote:Actually, this is not true. There are enough differences between the standard YL 120 fan and the Nexus 120 to make them quite different animals. I believe those differences are well documented in our review of the Nexus.... way back when in our of our first 120mm fan roundups.
My understanding is that the Yate Loons are all based on a shared reference design, with basically two types of blades (and probably some variations on that) and different electronics/electricals. Not unlike the family of Scythe Slipstreams and the family of Nidec Gentle Typhoon fans.
Perhaps this is true... but I have never encountered the much finer blade design of the Nexus in any YL120s acquired on the market.

ces
Posts: 3395
Joined: Thu Feb 04, 2010 6:06 pm
Location: US

Re: What I have Learned about Computer Performance from SPCR

Post by ces » Wed Feb 09, 2011 8:40 am

MikeC wrote: I have never encountered the much finer blade design of the Nexus in any YL120s acquired on the market.
I have never used Yate Loon. Why bother when you can get the Nexus for a few bucks more?

But a lot of people seem to like them. Petra appears to carry a model only available from Petras that sure looks a lot like the twisted Nexus fan blades... but maybe with a rougher finish. Though the slowest version of it is a 1300 rpm model as opposed to the quieter Nexus 1000 rpm model:

http://www.petrastechshop.com/12yalod1cafa.html

If you are looking for something that will put out some high CFM with decent static pressure, their 2200 rpm version might not be a bad choice over the 1900 RPM Slipstream, especially at Petra's current price of $5.99 for one and 4 for $20.00.
You do not have the required permissions to view the files attached to this post.

MikeC
Site Admin
Posts: 12285
Joined: Sun Aug 11, 2002 3:26 pm
Location: Vancouver, BC, Canada
Contact:

Re: What I have Learned about Computer Performance from SPCR

Post by MikeC » Wed Feb 09, 2011 9:02 am

ces -- that's exactly what I am talking about. The Nexus & YL fans you posted are not that close in performance, and they look quite different when examined closely. The YL is pretty good -- but it would never be our reference.

ame
Posts: 488
Joined: Tue Jul 10, 2007 10:35 pm
Location: Israel

Re: What I have Learned about Computer Performance from SPCR

Post by ame » Wed Feb 09, 2011 11:12 am

What I appreciate the most about SPCR is the high standards when it comes to making proper comparison between 2 essentially different products.

Be it noise level measurements, that are done using very sensitive equipment in a highly isolated environment. Testing cooling ability by using a 'one standard for all' reference fan (that is probably one of the best fans ever made). Pointing to the subtle differences in blade design and bearing/motor types and the way these factors affect noise and cooling.

These aspects are so often ignored by many on line reviewers, tilting the results of reviews and roundups one way or another in ways that are obviously flawed.

There is no doubt there is a lot that can be learned from SPCR's Mike, and the rest of the moderators and forum friends.

ces
Posts: 3395
Joined: Thu Feb 04, 2010 6:06 pm
Location: US

Re: What I have Learned about Computer Performance from SPCR

Post by ces » Wed Feb 09, 2011 12:06 pm

ame wrote:What I appreciate the most about SPCR is the high standards when it comes to making proper comparison between 2 essentially different products.
Yeah, there is a level of careful thoughtfulness at SPCR that you just don't see anywhere else. There are other sites that in any particular review may be accurate and thorough, but just never as thoughtful, useful or practical.

The xbits fan reviews come to mind. Good reviews, covering lots of fans, lots of metrics, detailed graphs... but at the end of it all, its a lot of useless (albeit impressive) sword motion. It's not information that I can use and rely on to make a purchasing decision... not like the information at SPCR.

At the end of the day, I don't find the xbits fan reviews of much use in making a good decision. You look at xbits rankings and there is something missing. You go to an SPCR product ranking, and you can rely on it... even when you don't fully understand the subtleties in the differences between products... maybe especially when you don't understand the differences. I have learned sometimes you just need to follow the SPCR recommendations and you figure out why later (a lesson that has cost me a few bucks more than once in the past).

I am really looking forward to the new forthcoming SPCR fan review methodology that they have been tinkering with for a while. I am hoping it is going to improve on what, in my personal opinion, are already the most thoughtful, penetrating and useful fan reviews on the internet.

alecmg
Posts: 204
Joined: Thu Mar 13, 2008 5:56 am
Location: Estonia

Re: What I have Learned about Computer Performance from SPCR

Post by alecmg » Thu Feb 10, 2011 2:01 am

ces wrote:
alecmg wrote:Btw Sandy Bridge processors break these rules in many aspects.
Please explain further.
First, its cache system and memory controller are very efficient. So size of L3 cache (LLC) does not play a major role anymore.
And effect of memory speed and timings is diminished.
Then Hyperthreading kicks in and blurs number of cores effect. i7-2600 and i5-2500 have 4 cores and similar speed, in encoding - very different.
And then SB are 20% faster per MHz ( I saw up to 80% faster compared to phenoms), can no longer be compared directly to other processors.

If you add to that its power efficiency, it makes Sandy Bridge best processor hands down.

ces
Posts: 3395
Joined: Thu Feb 04, 2010 6:06 pm
Location: US

Re: What I have Learned about Computer Performance from SPCR

Post by ces » Thu Feb 10, 2011 6:00 am

alecmg wrote:First, its cache system and memory controller are very efficient. So size of L3 cache (LLC) does not play a major role anymore.
And effect of memory speed and timings is diminished.
Then Hyperthreading kicks in and blurs number of cores effect. i7-2600 and i5-2500 have 4 cores and similar speed, in encoding - very different.
And then SB are 20% faster per MHz (I saw up to 80% faster compared to phenoms), can no longer be compared directly to other processors.
If you add to that its power efficiency, it makes Sandy Bridge best processor hands down.
Well this is all going to get very interesting very quickly. Apparently because of concern that the Atom isn't keeping up competitively in the "pad" marketplace, Intel is rushing Ivy Bridge out before the end of the year. Those will be even more energy efficient... presumably the dual core Ivy Bridge chips will be suitable for use in higher end pads.

ces
Posts: 3395
Joined: Thu Feb 04, 2010 6:06 pm
Location: US

Re: What I have Learned about Computer Performance from SPCR

Post by ces » Thu Feb 10, 2011 6:59 am

alecmg wrote:(I saw up to 80% faster compared to phenoms)
Do you have any links on this?

alecmg
Posts: 204
Joined: Thu Mar 13, 2008 5:56 am
Location: Estonia

Re: What I have Learned about Computer Performance from SPCR

Post by alecmg » Mon Feb 14, 2011 4:34 am

ces wrote:
alecmg wrote:(I saw up to 80% faster compared to phenoms)
Do you have any links on this?
here
Phenom II X4 at 3,6 vs i5-2500K at 3,3 plus turbo
Roughly same clocks and same number of cores and memory.
In gimp test its 80%, in other cpu-dependant tests consistently 40-50% faster

ame
Posts: 488
Joined: Tue Jul 10, 2007 10:35 pm
Location: Israel

Re: What I have Learned about Computer Performance from SPCR

Post by ame » Mon Feb 14, 2011 4:48 am

alecmg wrote:
ces wrote:
alecmg wrote:(I saw up to 80% faster compared to phenoms)
Do you have any links on this?
here
Phenom II X4 at 3,6 vs i5-2500K at 3,3 plus turbo
Roughly same clocks and same number of cores and memory.
In gimp test its 80%, in other cpu-dependant tests consistently 40-50% faster
It even beats the x6

goatsandmonkeys
Posts: 91
Joined: Wed Oct 06, 2010 6:32 am
Location: new york

Re: What I have Learned about Computer Performance from SPCR

Post by goatsandmonkeys » Mon Feb 14, 2011 4:06 pm

ces wrote:I am really looking forward to the new forthcoming SPCR fan review methodology that they have been tinkering with for a while. I am hoping it is going to improve on what, in my personal opinion, are already the most thoughtful, penetrating and useful fan reviews on the internet.
When might we see the new methodology and a new round up? Also can we expect a 14omm fan round up any time soon?

Also I've seen the Yate Joon's as low as 4-5 bucks and in the test they do EXTREMELY similar to the Nexus at low speeds. I use one for intake over my hard drives at a low speed and it's great and cheap.

ces
Posts: 3395
Joined: Thu Feb 04, 2010 6:06 pm
Location: US

Re: What I have Learned about Computer Performance from SPCR

Post by ces » Mon Feb 14, 2011 5:54 pm

goatsandmonkeys wrote:When might we see the new methodology and a new round up? Also can we expect a 14omm fan round up any time soon?.
January 2011 is that last date that I heard.

I don't care if it is behind schedule though. I think it is more important that they get it right than rush it.... and it is a squirrely balancing task to get it right. Sooner or later it will get done, and I am highly confident it will be worth the wait.

Post Reply