A computer components & hardware forum. HardwareBanter

If this is your first visit, be sure to check out the FAQ by clicking the link above. You may have to register before you can post: click the register link above to proceed. To start viewing messages, select the forum that you want to visit from the selection below.

Go Back   Home » HardwareBanter forum » Processors » General
Site Map Home Register Authors List Search Today's Posts Mark Forums Read Web Partners

VIA C3 cpu benchmark



 
 
Thread Tools Display Modes
  #1  
Old December 1st 03, 03:50 PM
terry
external usenet poster
 
Posts: n/a
Default VIA C3 cpu benchmark

Could anyone compare the speed of a VIA CPU (1GHz) comparing with
Celeron 1GHz and PIII-1GHz for database applications?

Is it PIIIVIACeleron or PIII about equal to VIA?

Thanks!
  #3  
Old December 2nd 03, 03:20 AM
Rob Stow
external usenet poster
 
Posts: n/a
Default

z wrote:

(terry) wrote in
m:


Could anyone compare the speed of a VIA CPU (1GHz) comparing with
Celeron 1GHz and PIII-1GHz for database applications?

Is it PIIIVIACeleron or PIII about equal to VIA?

Thanks!



dono if this will help but my c3 866 is pretty dog slow for FPU intensive
stuff. -- makes a nice little linux box and is uber quiet (and small mini-
atx).

If you need speed pIII is probably best of those choices


I've set a few dozen people up with C3 systems over the last
couple of years. If you want a fanless system or a PSU-fan
only system that is good enough to run XP, web browser, word
processor, play DVD's, etc, then a 1 GHz C3 is more than good
enough for most people. Other than that - forget it, a C3 is
almost undoubtedly not for you.

Once you get outside of the basic tasks I outlined above, a
pre-Nehemiah C3 is like a Celeron or Duron at 1/3 the clock
speed, while a Nehemiah does a little better. If you think
a 500 MHz Celery or Duron would be good enough for your database
app, then I suppose a 1 GHz Nehemiah would be good enough as
well. I suppose it all depends on how big the database is,
how many C3 systems you are prepared to cluster together, how
many users, ...

Also be sure to use avoid VIA's PLE133 chipset.
Their CLE266 chipset, however, is quite a big improvement
and I would recommend it.


--
Reply to newsgroup only please. This e-mail account is real
but effectively abandoned because of excessive spamming.
  #4  
Old December 2nd 03, 03:21 PM
Tony Hill
external usenet poster
 
Posts: n/a
Default

On Mon, 01 Dec 2003 16:36:55 -0600, z wrote:
(terry) wrote in
om:

Could anyone compare the speed of a VIA CPU (1GHz) comparing with
Celeron 1GHz and PIII-1GHz for database applications?

Is it PIIIVIACeleron or PIII about equal to VIA?

Thanks!


dono if this will help but my c3 866 is pretty dog slow for FPU intensive
stuff.


Yeah, and databases applications are real FPU intensive (in a
not-at-all kind of way).

Database applications tend to require three things, cache, low
latency/high bandwidth memory (whether latency or bandwidth is more
important usually depends on what sort of operations you do mostly)
and the speed of I/O and storage. Assuming that the last is equal
between systems, you see that the VIA C3 has a larger L1 cache (64K
instruction + 64K data) but smaller L2 cache (64K) as compared to a
Celeron (16K/16K L1 + 128K L2). In total the VIA C3 can cache up to
128K of data though (it uses a victim cache design for the L2, unlike
the Celeron), essentially the same as the Celeron. The larger 256K
cache of the PIII tends to make it a noticeably quicker on database
apps when compared to the Celeron.

Both chips use the same GTL bus, though the VIA C3 uses a 133MHz bus
speed vs. the 100MHz bus speed of the Celeron, so this should give an
edge to the C3. The PIII also uses a 133MHz bus.

For the actual processor itself, this usually doesn't factor into
databases TOO much, though it will have an effect. Here the Celeron
will probably be a bit faster, the C3 was really designed with
low-cost/low power in mind, so some ops are slower. For databases
though, most of the ops used will tend to be fairly standard
load/stores, maybe a few adds, multiplies, etc. and branches. None of
these are very slow on the C3, so it should be pretty competitive with
the Celeron.

In total, I would through out a guesstimate that the VIA C3 will
probably be about 5-10% slower than a similarly clocked Celeron for
database stuff, and the PIII being 10-15% quicker than the Celeron.
Of course, I'd give my estimate at least a 25% margin of error! :

If anyone has some real-world numbers, please feel free to share!

-------------
Tony Hill
hilla underscore 20 at yahoo dot ca
  #6  
Old December 2nd 03, 07:54 PM
z
external usenet poster
 
Posts: n/a
Default

Tony Hill wrote in
.com:

On Mon, 01 Dec 2003 16:36:55 -0600, z wrote:
(terry) wrote in
. com:

Could anyone compare the speed of a VIA CPU (1GHz) comparing with
Celeron 1GHz and PIII-1GHz for database applications?

Is it PIIIVIACeleron or PIII about equal to VIA?

Thanks!


dono if this will help but my c3 866 is pretty dog slow for FPU
intensive stuff.


Yeah, and databases applications are real FPU intensive (in a
not-at-all kind of way).


right. the FPU being slow is the only thing i've noticed as a negative
of this chip in using it for the last year or so is all.
  #7  
Old December 2nd 03, 08:47 PM
Rob Stow
external usenet poster
 
Posts: n/a
Default

z wrote:

Tony Hill wrote in
.com:


On Mon, 01 Dec 2003 16:36:55 -0600, z wrote:

(terry) wrote in
.com:


Could anyone compare the speed of a VIA CPU (1GHz) comparing with
Celeron 1GHz and PIII-1GHz for database applications?

Is it PIIIVIACeleron or PIII about equal to VIA?

Thanks!


dono if this will help but my c3 866 is pretty dog slow for FPU
intensive stuff.


Yeah, and databases applications are real FPU intensive (in a
not-at-all kind of way).



right. the FPU being slow is the only thing i've noticed as a negative
of this chip in using it for the last year or so is all.


The FPU situation is much improved with the Nehemiah C3's
compared to the pre-Nehemiah versions, but still not on a
par with the Celery or Duron.



--
Reply to newsgroup only please. This e-mail account is real
but effectively abandoned because of excessive spamming.
  #8  
Old December 3rd 03, 04:39 PM
Evgenij Barsukov
external usenet poster
 
Posts: n/a
Default

Tony Hill wrote:

On Mon, 01 Dec 2003 16:36:55 -0600, z wrote:
(terry) wrote in
om:

Could anyone compare the speed of a VIA CPU (1GHz) comparing with
Celeron 1GHz and PIII-1GHz for database applications?

Is it PIIIVIACeleron or PIII about equal to VIA?

Thanks!


dono if this will help but my c3 866 is pretty dog slow for FPU intensive
stuff.


Yeah, and databases applications are real FPU intensive (in a
not-at-all kind of way).

Database applications tend to require three things, cache, low
latency/high bandwidth memory (whether latency or bandwidth is more
important usually depends on what sort of operations you do mostly)
and the speed of I/O and storage. Assuming that the last is equal
between systems, you see that the VIA C3 has a larger L1 cache (64K
instruction + 64K data) but smaller L2 cache (64K) as compared to a
Celeron (16K/16K L1 + 128K L2). In total the VIA C3 can cache up to
128K of data though (it uses a victim cache design for the L2, unlike
the Celeron), essentially the same as the Celeron. The larger 256K
cache of the PIII tends to make it a noticeably quicker on database
apps when compared to the Celeron.

Both chips use the same GTL bus, though the VIA C3 uses a 133MHz bus
speed vs. the 100MHz bus speed of the Celeron, so this should give an
edge to the C3. The PIII also uses a 133MHz bus.

For the actual processor itself, this usually doesn't factor into
databases TOO much, though it will have an effect. Here the Celeron
will probably be a bit faster, the C3 was really designed with
low-cost/low power in mind, so some ops are slower. For databases
though, most of the ops used will tend to be fairly standard
load/stores, maybe a few adds, multiplies, etc. and branches. None of
these are very slow on the C3, so it should be pretty competitive with
the Celeron.

In total, I would through out a guesstimate that the VIA C3 will
probably be about 5-10% slower than a similarly clocked Celeron for
database stuff, and the PIII being 10-15% quicker than the Celeron.
Of course, I'd give my estimate at least a 25% margin of error! :

If anyone has some real-world numbers, please feel free to share!


Here goes:
http://www6.tomshardware.com/cpu/20020605/

Basic idea is - Via C3 1GHz is slower then Celeron 800 in most tests, but considering
how chip and heat-saving it is, still a good deal.

Regards,
Yevge



--

__________________________________________________
*science&fiction*free programs*fine art*phylosophy:
http://sudy_zhenja.tripod.com
----------remove hate_spam to answer--------------
  #9  
Old December 3rd 03, 07:37 PM
Rob Stow
external usenet poster
 
Posts: n/a
Default

Evgenij Barsukov wrote:

Tony Hill wrote:

On Mon, 01 Dec 2003 16:36:55 -0600, z wrote:

(terry) wrote in
.com:


Could anyone compare the speed of a VIA CPU (1GHz) comparing with
Celeron 1GHz and PIII-1GHz for database applications?

Is it PIIIVIACeleron or PIII about equal to VIA?

Thanks!


dono if this will help but my c3 866 is pretty dog slow for FPU intensive
stuff.


Yeah, and databases applications are real FPU intensive (in a
not-at-all kind of way).

Database applications tend to require three things, cache, low
latency/high bandwidth memory (whether latency or bandwidth is more
important usually depends on what sort of operations you do mostly)
and the speed of I/O and storage. Assuming that the last is equal
between systems, you see that the VIA C3 has a larger L1 cache (64K
instruction + 64K data) but smaller L2 cache (64K) as compared to a
Celeron (16K/16K L1 + 128K L2). In total the VIA C3 can cache up to
128K of data though (it uses a victim cache design for the L2, unlike
the Celeron), essentially the same as the Celeron. The larger 256K
cache of the PIII tends to make it a noticeably quicker on database
apps when compared to the Celeron.

Both chips use the same GTL bus, though the VIA C3 uses a 133MHz bus
speed vs. the 100MHz bus speed of the Celeron, so this should give an
edge to the C3. The PIII also uses a 133MHz bus.

For the actual processor itself, this usually doesn't factor into
databases TOO much, though it will have an effect. Here the Celeron
will probably be a bit faster, the C3 was really designed with
low-cost/low power in mind, so some ops are slower. For databases
though, most of the ops used will tend to be fairly standard
load/stores, maybe a few adds, multiplies, etc. and branches. None of
these are very slow on the C3, so it should be pretty competitive with
the Celeron.

In total, I would through out a guesstimate that the VIA C3 will
probably be about 5-10% slower than a similarly clocked Celeron for
database stuff, and the PIII being 10-15% quicker than the Celeron.
Of course, I'd give my estimate at least a 25% margin of error! :

If anyone has some real-world numbers, please feel free to share!



Here goes:
http://www6.tomshardware.com/cpu/20020605/

Basic idea is - Via C3 1GHz is slower then Celeron 800 in most tests, but considering
how chip and heat-saving it is, still a good deal.


Those THG numbers are out of date and totally irrelevant - they
predate the C3-Nehemiah. If you buy a 1 GHz C3 today it will
probably be a Nehemiah and a Nehemiah is a good improvement over
the pre-Nehemiah C3's.


Just found this review:
http://www.digit-life.com/articles2/...-nehemiah.html
It compares
- 1 GHz Nehemiah, VIA CLE266 chipset
- 1 GHz Nehemiah, Intel i815P chipset
- 1.3 GHz Celery/Tualatin
Which makes me reconsider my recommendation of the CLE266 chipset
unless you intend to play DVDs a lot: the i815P beats the CLE266
in everything else, often by very large margins.

And I believe "Whining Dog" compared a C3-Nehemiah
and a pre-Nehemiah against a 2.x GHz P4.


Note also that the 1.3 GHz Nehemiah should be available now or very
soon. There are also rumours that a C3 version that is pin-compatible
with a P4 is under development - which would mean you will no longer
have to settle for a Socket 370 motherboard.



--
Reply to newsgroup only please. This e-mail account is real
but effectively abandoned because of excessive spamming.
  #10  
Old December 3rd 03, 11:29 PM
Tony Hill
external usenet poster
 
Posts: n/a
Default

On Wed, 03 Dec 2003 13:37:20 -0600, Rob Stow
wrote:
In total, I would through out a guesstimate that the VIA C3 will
probably be about 5-10% slower than a similarly clocked Celeron for
database stuff, and the PIII being 10-15% quicker than the Celeron.
Of course, I'd give my estimate at least a 25% margin of error! :

If anyone has some real-world numbers, please feel free to share!



Here goes:
http://www6.tomshardware.com/cpu/20020605/

Basic idea is - Via C3 1GHz is slower then Celeron 800 in most tests, but considering
how chip and heat-saving it is, still a good deal.


Those THG numbers are out of date and totally irrelevant - they
predate the C3-Nehemiah. If you buy a 1 GHz C3 today it will
probably be a Nehemiah and a Nehemiah is a good improvement over
the pre-Nehemiah C3's.

Just found this review:
http://www.digit-life.com/articles2/...-nehemiah.html



Actually the numbers on this test are every bit as irrelevant as the
THG numbers for the original poster. Let's see, they have:

CPU RightMark: FPU/SSE test
MP3 encoding: FPU/memory test
MPEG encoding: FPU/memory test
ACE archiving: ??? I'd have thought it's a disk-intensive test
3DMax Rendering: FPU test
Sysmark: System wide test with no way to get individual scores
3DMark: Mainly a video card test
Games: FPU/video card test

So how, exactly do any of these tests remotely demonstrate how the
chip will perform on database applications? It's a well established
fact that the chip isn't so hot at FPU stuff, but database use
basically ZERO floating point instructions. Reviews that are full of
FPU tests aren't overly useful to the original poster.

It compares
- 1 GHz Nehemiah, VIA CLE266 chipset
- 1 GHz Nehemiah, Intel i815P chipset
- 1.3 GHz Celery/Tualatin
Which makes me reconsider my recommendation of the CLE266 chipset
unless you intend to play DVDs a lot: the i815P beats the CLE266
in everything else, often by very large margins.


That's mainly due to the integrated video. The video for this chipset
doesn't contain any of the bandwidth-saving technologies of Intel's
chipset, and performance of integrated video is 99% determined by
memory bandwidth.

-------------
Tony Hill
hilla underscore 20 at yahoo dot ca
 




Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

vB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Forum Jump

Similar Threads
Thread Thread Starter Forum Replies Last Post
Suggestions for planned bonnie++ benchmark Jochen Kaiser Storage & Hardrives 0 April 13th 04 02:01 PM
REQ: benchmark help H.W. Stockman Overclocking AMD Processors 0 April 12th 04 07:07 PM
Athlon 64 Vs. Pentium 4 article: On the Justification for Quake3 as a CPU Benchmark rms Overclocking AMD Processors 7 October 5th 03 10:05 PM
Athlon 64 Vs. Pentium 4 article: On the Justification for Quake3 as a CPU Benchmark rms Overclocking 6 October 2nd 03 05:16 PM
Athlon 64 Vs. Pentium 4 article: On the Justification for Quake3 as a CPU Benchmark rms General 6 October 2nd 03 05:16 PM


All times are GMT +1. The time now is 02:58 PM.


Powered by vBulletin® Version 3.6.4
Copyright ©2000 - 2024, Jelsoft Enterprises Ltd.
Copyright ©2004-2024 HardwareBanter.
The comments are property of their posters.