1.6ghz centrino = 3ghz pentium?

fbrdphreak

Lifer
Apr 17, 2004
17,555
1
0
Equal to? No, they're very different processors with strenghts & weaknesses.

In terms of overall performance, then I would say a 1.6GHz Pentium M would be similar to a 2.8GHz Pentium 4. But again, they have strengths and weaknesses.
 

Hacp

Lifer
Jun 8, 2005
13,923
2
81
Originally posted by: andrewbabcock
No it equals more than that. 3.2 ghz p4 imo. P-M can do more calculations per clock than any processor on the market and that is why it is best. Runs very cool, with very little power, and is the fastest processor /clock on the market.

Wrong. A 1.6 GHZ pM should be around a 2.8 GHZ P4 in terms of performance OVERALL. There are defenetly desktop applications where the P-M falls down dying. There are some where it wins by 2 miles. Look at anandtech's review of the Pentium M for full analysis. You will see that a 2.0GHZ Pentium M generally compares equally with a 2.0 AMD 64.
 

fbrdphreak

Lifer
Apr 17, 2004
17,555
1
0
Originally posted by: andrewbabcock
No it equals more than that. 3.2 ghz p4 imo. P-M can do more calculations per clock than any processor on the market and that is why it is best. Runs very cool, with very little power, and is the fastest processor /clock on the market.
I've seen you spout this in other threads also. I don't mean to be rude, but you really don't know what you are talking about. Please stop
 

cheesehead

Lifer
Aug 11, 2000
10,079
0
0
If you're running "office" programs like Excel, Word, FrontPage, and the like, a P-M is a wonderful processor. I don't own one, but a 1.3 ghz Celeron M (essentally a Pentium M with less cache and disabled power management) seems much faster in these sort of programs than my 3ghz P4 laptop, which also has quadruple the memory of the celly-M laptop and vastly more graphics card performance.
A 1.6ghz P-M is maybe equivalent to a 2.6 P4 in some areas (mostly heavy multimedia that utilized multithreading), though you'll see performance vary wildly.
All things considered, The pentium M seems to be the analouge of the original V.W. bug: it has an absurdly small engine (in this case, essentially an updated Pentium Pro), but it's so tiny that it really does'nt need one. It's hardly a Mustang, but it's a pretty good car for driving about town nonetheless.
 

bearxor

Diamond Member
Jul 8, 2001
6,605
3
81
When I talk about PentiumM performance to someone interested in clock speeds, I give myself a 400mhz margin of error. I would say that a PentiumM 1.6 (533) falls somewhere inbetween a 2.6 and 3.0ghz Pentium4, depending on what you're doing. I then overlap that with higher processors.

.Pm.......P4...
1.60 - 2.6/3.0
1.73 - 2.8/3.2
1.86 - 3.0/3.4
2.00 - 3.2/3.6
2.13 - 3.4/3.8
2.26 - 3.6/4.0

That is assuming of course, 533mhz, 2MB L2 cache chips. I'd probably knock off another 400mhz for the older PentiumM's, 1MB L2 cache 400mhz FSB. That means I would put a 1.6ghz CeleronM (ones selling currently) at about a 2.4/2.8ghz Pentium4.

The truth is, you can't really compare a PentiumM to a Pentium4. You can pick a clock speed, but the problem is, the PentiumM is going to be faster at some things and slower at others. I've seen benchmarks all over the map. That's why I make perfectly clear that 'It is about the speed of a P4 2.6 to a 3.0. I'm pretty certain that should cover the performance of where a vast majority of benchmarks put the respective processors.

Truth is, I've completely gone away from speed and focused solely on processor numbers. It helped my sales out, and I only talked actual frequency with so-called 'geeks' that I could probably spend 5 minutes writing random stuff on a paper and have more than they have collected their entire lives, but I digress. In the end, it's all about how it feels to you, and honestly, I can't tell the difference between my 1.6 (533) processor in my 6000D and the 1.6 (400) that I swapped and then pinmodded to 2.13ghz, but thats just me.
 

Smbu

Platinum Member
Jul 13, 2000
2,403
0
0
Originally posted by: bearxor


That is assuming of course, 533mhz, 2MB L2 cache chips. I'd probably knock off another 400mhz for the older PentiumM's, 1MB L2 cache 400mhz FSB. That means I would put a 1.6ghz CeleronM (ones selling currently) at about a 2.4/2.8ghz Pentium4.
So what you say for the older Dothan P-M's? the 400mhz fsb, 2MB L2 chips? Just curious.
 

cheesehead

Lifer
Aug 11, 2000
10,079
0
0
Actually, I was thinking of the older ones.
(Dang, am I out of date.)
Yah, 2.8ghz eqivalent sounds about right.
 

andrewbabcock

Senior member
Oct 2, 2005
561
0
0
Originally posted by: fbrdphreak
Originally posted by: andrewbabcock
No it equals more than that. 3.2 ghz p4 imo. P-M can do more calculations per clock than any processor on the market and that is why it is best. Runs very cool, with very little power, and is the fastest processor /clock on the market.
I've seen you spout this in other threads also. I don't mean to be rude, but you really don't know what you are talking about. Please stop

Alright but P-M is still the best processor on the market. I don't care.

 

dexvx

Diamond Member
Feb 2, 2000
3,899
0
0
Get an Asus CT-479 adapter with a 865/875 board.

Overclock the Pentium-M to 2.4+ Ghz at 200+ FSB and it will really fly.
 

bearxor

Diamond Member
Jul 8, 2001
6,605
3
81
The 2mb 400mhz aren't MUCH slower than equivalent 533mhz processors. 200mhz slower? I think that's probably too much though. You could compare them to older 533mhz fsb P4's. Like 2.53, 2.66, 2.8, 2.93, 3.06, etc, etc. 533mhz FSB didn't seem to do alot overall for PentiumM, it was more of a complete platform overhaul than anything else.
 

Ronin

Diamond Member
Mar 3, 2001
4,563
1
0
server.counter-strike.net
Originally posted by: andrewbabcock
Originally posted by: fbrdphreak
Originally posted by: andrewbabcock
No it equals more than that. 3.2 ghz p4 imo. P-M can do more calculations per clock than any processor on the market and that is why it is best. Runs very cool, with very little power, and is the fastest processor /clock on the market.
I've seen you spout this in other threads also. I don't mean to be rude, but you really don't know what you are talking about. Please stop

Alright but P-M is still the best processor on the market. I don't care.

Might want to look into a Turion, buddy. ;)
 

First

Lifer
Jun 3, 2002
10,518
271
136
A 1.6GHz 1MB L2 400MHz FSB P-M is roughly equal to a 2.7GHz P4. A 1.6GHz 2MB L2 533MHz FSB P-M is roughly equal to a 2.9GHz P4.

EDIT: Assuming we're talking about the 90nm Prescott P4's. The Pentium D's are a different story, somewhat.
 

cheesehead

Lifer
Aug 11, 2000
10,079
0
0
Turions are wonderful processors, too. It's really like choosing between two sportscars: you'll be happy either way. I personally would get the Turion, but that's because I like my laptops rip-roaring fast and don't want to pay for DDR2. The Pentium-M Sonoma platform, however, performs at least as well.
Either way, you can't lose!
 

ND40oz

Golden Member
Jul 31, 2004
1,264
0
86
Originally posted by: bearxor
The 2mb 400mhz aren't MUCH slower than equivalent 533mhz processors. 200mhz slower? I think that's probably too much though. You could compare them to older 533mhz fsb P4's. Like 2.53, 2.66, 2.8, 2.93, 3.06, etc, etc. 533mhz FSB didn't seem to do alot overall for PentiumM, it was more of a complete platform overhaul than anything else.

I tend to disagree with this statement. My Dell D610 with the 1.73 533fsb dothan kills my X41 with a 1.5 400fsb dothan. The ULV X41 does use quite a bit less power, but it has no where near the performance of the dell. But my dell weighs over twice as much as well.
 

bearxor

Diamond Member
Jul 8, 2001
6,605
3
81
Yea, but you're also talking about a nice clock speed difference. If your X41 had a 1.7 400mhz, it likely wouldn't be very noticeable. My wife notices a difference between the 1.6 and the 1.8 I put in, but she says its not major. It's also entirely subjective. Like I said, I don't notice enough of a difference in between the 1.6 I had in here to begin with and the 1.6 that I have overclocked to 2.13.

That's why I give more of a range to customers instead of a dead set 'it's equal to a P4 2.8'.
 

Pabster

Lifer
Apr 15, 2001
16,986
1
0
Overall they are pretty close.

For example, my Dothan 1.86 runs a bit faster than my 3.0 630 Prescott. But not in every single task, just in general.

If we're talking power consumption the Dothan has a huge advantage, of course.
 

fbrdphreak

Lifer
Apr 17, 2004
17,555
1
0
Originally posted by: Cheesehead
Turions are wonderful processors, too. It's really like choosing between two sportscars: you'll be happy either way. I personally would get the Turion, but that's because I like my laptops rip-roaring fast and don't want to pay for DDR2. The Pentium-M Sonoma platform, however, performs at least as well.
Either way, you can't lose!
Actually DDR2 is quite cheap now and is getting cheaper. AMD is moving to DDR2 with their new platforms in 2006, and video card mfrs are moving to DDR2 as the base memory for GPU's. DDR2 should drop nicely in price Q1 06 :thumbsup:
 

cheesehead

Lifer
Aug 11, 2000
10,079
0
0
Yah, but the difference is not too big unless you're running 667mhz RAM. That said, if I have to get DDR2 to get a good graphics card, I'm getting DDR2.
 

arcas

Platinum Member
Apr 10, 2001
2,155
2
0
Example of a real-world benchmark:

I wrote a stock analysis/simulation program that I run daily. It manipulates between 100MB and 1.2GB data in RAM though at any given time it's operating on a small subset of that data so it's not as memory intensive as, say, a fluid dynamics simulation.

Back in July, I ran some comparisons between the machines at my disposal (each running Linux, stock program was compiled with gcc 3.4 optimized for the particular processor in use).

On my 1.4GHz T-bird PC133, it ran in about 90 minutes
On my 2.0GHz P-M (400mhz bus), it ran in about 25 minutes
On a dual 3.2GHz IBM xServer 330 (I think), running on a single CPU, it was also about 25 minutes. I remember it being just a touch faster than the P-M but close enough to get lost in the noise. Granted, had I built the application to run 2 threads, the x330 would have blown the Thinkpad out of the water but that's to be expected since the app throughput scales linearly with the number of threads.

I must admit the T-bird only had 512MB of RAM while the other 2 machines each had 2GB RAM (yes, my Thinkpad has 2GB :) )but this particular dataset was only about 280MB so no swapping took place.