Measure your Dual/Multi Core Notebook CPU Speed - Bye Bye to SuperPI

Fired up my older notebook and here's what I got:

32M: 64.891 seconds
1024M: 35.1638 minutes

Compaq R4000
Athlon64 3800+ @ 2.4GHz (single-core)
768MB DDR333
Windows Vista Ultimate

 

Asus W2V - Windows Media Center / SP2
Dothan 2.13ghz
84.875 sec

 

That score seems inaccurate, it should be half that time.

Remember to make sure that
- you plug your notebook into AC Power (to give maximum performnce)
- wPrime is set to 2 Threads in the Advanced Settings.

And make sure that you have configured your XP to fully support multi-core CPUs by doing the XP hotfix

EDIT: Sorry didnt see that it was a single core Pentium M

 

R4000,
Thats a nice single core score.

Why is this your old notebook? It seems pretty nice, except it can use a bit more RAM... then it would be a nice all-around notebook. Then it would beat the Sempron system that you have.

 

Yeah, aside from the clunkier design and fan noise it is a superior performing notebook to the V5201. It even has the 128MB dedicated vram x200m, which is leaps better than the shared version in my present lappy. However, the 130nm Newcastle cpu in it runs pretty hot compared to my newer 90nm Albany (Sempron).

Anyway, the R4000 was a test pig for Vista during the beta/release candidate stages. When Vista went RTM, the stock ram was removed and donated to other machines. I never even bothered to put XP back on it and just let the machine sit idle for months. I only fired it up with the junky "filler" ram to run your benchmark test.




Mike

 

LOL. I guess the heat and fan noise would be a consideration to move onto another notebook. But for "junky filler" RAM, it still gave a good benchmark.

 

Hey guys,

I appreciate all of the participation.

I am working on a new table ranking system, so if you post your score if you can.

Free REP for those that participate [or already have participated].

 

It appears to me that so far everyone has reported a notebook's time but since my notebook model and relative time has already been posted I thought I would run this test on my P4 custom desktop just to give you an idea how far we have come from the P4 Ghz race days.

Processor: Intel Pentium 4 (Prescott core) 3.2Ghz HT
Time: 90.14sec (this is the slowest time yet!)
Windows XP Professional

WOW compared to some of the other times posted here hyperthreading does not make running two processes fast, this processor seemed to struggle compared to my turion x2. IMO dual cores will blow away HT any day!!!

 

Thanks for posting, although that score might not be fully accurate.. especially since you are using XP (some mods needed to give your CPU full performance)

Remember to make sure that
- you plug your notebook into AC Power (to give maximum performnce)
- wPrime is set to 2 Threads in the Advanced Settings.
- (if using XP 32-bit w/ Service Pack2) make sure that you have configured your XP to fully support multi-core CPUs by doing the XP hotfix (install XP hotfix, registry add-on, and BOOT.ini add-on)

 

Gophn you seemed to misunderstand my post, that time was from my P4 (Prescott core) desktop so the plug in part was not applicable.
I also made sure that wPrime was set to 2 threads, however I will do the test again after applying the dual core fix but I doubt it will improve that time by much (I applied that fix to my aspire 5102 several months ago but never got around to doing it on the desktop ).

I gave this time to show that hyperthreading (simulated dual core) and the Pentium 4s ageing/ dead NetBurst core architecture are just no match for modern true dual core processors, but I think you only want notebooks in this thread so this wPrime time perhaps is irrelevant...

 

saying Prescott does not mean that its only desktop, I have a couple of friends with that CPU in their notebooks (Clevo beasts).

In any case, thanks for the quick replies.

 

Fastest time yet.

Thanks VirtueTech for participating.

 

Anyone else think its funny that the 2 fastest laptops so far are Thinkpads, and not "gaming" machines?

 

its just CPU benchmarking

And i just noticed that too, LOL.

Just wait until someone posts the new Clevo D900C (with a desktop Core 2 Duo/Extreme).

 

I'm glad that SuperPI is finally being put to pasture... with the widespread availability of multi-core CPUs it was becoming less and less relevant as an accurate benchmark.

This was brought to a head with the release of the T7 series Core 2 Duos with their 4MB L2 cache. Being an unrealistically cache-dependant app, we were seeing ridiculous apparent performance increases versus the equivalently clocked Core Duo CPU.

 

Dual booting between XP SP2 and XP x64 SP2. The regular XP is 4 seconds faster. 53 seconds to 57 seconds on multiple tests. Anyone know why? I thought x64 is supposed to be the faster OS?

my processor is a Turion 64 X2 with 2GB 667

 

Tested under XP SP2, right after dusting out my baby

 

If I remember correctly, having a 64bit CPU is only beneficial when you use the 64bit application, which will only run under 64bit OSs. I'm sure this test program only regards 32bit processing (as it runs under 32bit OSs), so indeed it should not distinguish between the 32bit and 64bit CPUs.

 

Here's the result against which the others can be judged.

The CPU is the 633MHz Crusoe with its legs tied together by 128MB RAM and Windows ME.

608.792s for 32M means just over 10 minutes while 21332.352s for 1024M is 5.92 hours.

John

 

All tables updated.

John, that Crusoe is a beast.

 

Is a snail a beast? It didn't seem that slow when I bought it! I think the other computers have got faster. But all that hard work didn't make the fan come on (because there isn't one).

John

PS: It is not in a desktop but a Fujitsu P2020 notebook

 

Ahh.. I see. Just updated it.

Definitely a notebook with a score that most cant match.

 

My score is 52.75s, see sig for spec and pic for proof.

BTW i did the CPU fix from your other thread including updating boot.ini and i got the same time in this and both super pi. The time in Wprime only changed by like 0.02 s!!

 

got it, looks to be the right score for that CPU.

 

Just to let you guys know, but wprime is not a very good stability test either.

My Dual 5120 server at 2700mhz can run the 32m calculation in almost less than 16 seconds and takes just under 500 seconds to run the 1024m calculation (old scores on hwbot at the moment).

What would be nice is a loop option.

Systool (which has pi benching built in) has this option.


All that said, my T7400es gets: http://www.hwbot.org/user.do?userId=5266

32m - 40s 20ms http://www.hwbot.org/result.do?resultId=570343
1024m - 21m 13s 250ms http://www.hwbot.org/result.do?resultId=570336

 

Well, my Single-Core P4 2.6 GHz got 169.705s on 32m. I'm too afraid it will break if I try 1024 .

Specs are (Desktop)
Dell Dimension 8300
P4 2.6 GHz
XP Home SP2
1 GB RAM
Radeon 9800 PRO 128 MB

Wow, I can't wait for my G1s.

 

btw... i tried to run this wprime on my notebook with Vista Home Premium German version... but i got an error msg, "unknown error;quitting"... but i tried to run it on my desktop with XP Home Edition SP2... its running without problem...

is it because my vista installation is faulty?

 

did you try the console version instead?

http://www.wprime.net/wPrime_r142.zip

 

not yet... okay... will give a go with the console... cause i was not sure if i should try the other version... thanks for the quick reply Gophn...

 

STEvil and Mr._Kubelwage, thanks for the participation.

Hall of Fame updated.

 

btw.. the problem settled after i choose to run the program as administrator... then i can run it normally. without a problem. so here is my result...

Samsung Q35 Core 2 Duo T5500 1.66Ghz | Vista Home Premium
32M 50,264secs
1024M 1597,687secs

and just some extra from my old desktop...

FS scaleo 600 Pentium 4 3.2Ghz HT | XP SP2
32M 89,359secs
1024M 3160,187secs

 

Got it and updated the hall of fame.

Thanks for the extra info on the Vista admin issue with this benchmark.

 

here's mine
32M 70.54secs
1024M 2450secs

Aspire 5102AWLMi
AMD Turion 64 mobile technology MK-38 (2.2 GHz, 512kb L2 cache)
15.4" WXGA Acer CrystalBrite LCD(16ms)
Up to 384MB ATI Mobility Radeon X1300 HyperMemory
120 GB HDD
DVD-Super Multi double layer(Support DVR+/- Double Layer/DVR+/-RW)
1GB DDR2
802.11B/G WIRELESS LAN
OS - Window Vista Home Premium

 

Additional info... the admin account is needed on XP Profesional too...

cause I try to run this software on PC at my office, using normal user account... same error.

So Admin right is needed to run this benchmark on XP n Vista... just some info for others..

 

Thanks for posting your scores.

That AMD Turion seems like a great CPU... especially for the supposed inexpensive notebooks.

Thanks for the info, i'll note it on the thread.

 

your welcome m8

 

can I join the list?

32M 30.265 sec
1024M 955.266 sec

Sager 9750

AMD [email protected]/Vista Ultimate x64

 

nice!! AMD FX-60 FTW!

Finally a true high-end notebook with all the bells and whistles.

 

nice result... impressive... I guess this is the best result for now is it?

 

Yup this AMD FX-60 score just blew away the competition.

Leave it to Clevo to use desktop CPU's in their notebooks to yield such power.

 

maybe this is lil bit off the topic here... but by mean... currently only clevo manufactured notebook why such high processor capabilities is it?
interesting.... cause such notebooks can be such powerfull...

 

The Clevo Dxxx series have been infamous for all being desktop replacement models that use desktop CPUs.

But the main downside is the batterylife. With a 12-cell in these models, the average batterylife is 1.3-1.5 hours.

The newest D900C [w/ desktop Core 2 Duo/Extreme] should blow away the D900K's score [w/ Athlon FX-60].

I am just waiting for the benchmarks.

I believe that a few other ODMs (like Asus) are about to release some models using desktop CPUs... but Clevo has pretty much cornered that market niche... thats why they have been known to make the craziest notebooks.

 

Its still better than the 45 minutes I got with a 12-cell on my A64 3800+ desktop replacement lappy (R4000). Horrible battery life.........

 

LOL. I get that time if I am trying to game.

These desktop replacements are definitely great for power on the go, but with dismal batterylife.

 

Uh, oh I smell trouble! Gophn's wrong!?!?!

Just playin' with you

Again, a nice piece of work put together by Gophn.

 

The D900C score is up w/ E6700. 2nd top score... for the moment.

It is slightly slower than the FX-60, but I think it should be near the same if the benchmark was done in Vista.

I know that if someone sticks a X6800 in the D900C, that score should be #1.

 

Vista doesnt help Wprime, it is not coded in 64-bit.

Thats one heck of a fast FX-60, beats all the others on hwbot.org !

 

Well, that's not a fair fight? That just steals the fun out of competition, comparing two processors from different time frames, and knowing who's the victor.

 

I got ~27 mins in stress test, see pic.

 

Thanks for the new stress test score.

Hall of fame is updated.