• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

Processors and DC

O

Overkiller

Platinum Member
Hello,

most DC clients rely mostly on the processor and not much else (i.e. only use up 10 mb of ram and little to no of the hard drive)...

my question therefore lies in which is the best processor for DC clients?

particualrly something like SOB.

Does anyone have a link or can personally discuss the difference between Via's (hehe), AMD's, and Intels high end processors.
-Does the barton compete w/ the 3.06 ghz HT P4?

thanks guys!

dont make this a flame war. i was just curious which architecture can run a DC client better (or if its really that close)
 
Well, a high-end P4 on a very high FSB and running memory also on a very high speed is a good place to start 🙂

I know that.
 
I don't know if anyone has already done some tests with Barton's...

With a P4 you can't go wrong 😉

insteady of a 3.06, why not try a 2.8 or 2.66 and give it a bash in overclocking ?

a 2.8@3360 is lot faster than a 3.06@3450 and cheapper 😉
 
Originally posted by: Stormgiant
I don't know if anyone has already done some tests with Barton's...

With a P4 you can't go wrong 😉

insteady of a 3.06, why not try a 2.8 or 2.66 and give it a bash in overclocking ?

a 2.8@3360 is lot faster than a 3.06@3450 and cheapper 😉
Click to expand...
P4s are good in SETI, if combined with Rambus or dual channel DDR, AFAIK, but in most other projects, they suck in comparison with Athlons. This difference is particularly noticable in RC5, ECCp/2, and Folding@Home (the only three I have seen many benches for). Aside from that, Athlons have a good price/performance ratio (consider my 1700+ TbredB that cost US$50 and almost makes it to 2GHz, 2400+).
 
From what I've heard HT really doesn't do much at all for any DC project since it just does the same thing over and over and therefore ends up competing with resources. However, I have heard that HT does provide some speed increase if you run one instance of the client for each logical processor (hardly worth the money though in my oppinion.

Barton should provide a little bit of a performance increase due to a larger chunk of the client being able to fit in the cache but once again it would be minimal and should only scale a little better than the TBred did.

I woud say that with either processor your main bottle neck would probably be your FSB. FSB is where the P4 has an advantage. P4 seems to be the way to go with SoB in my oppinion.
 
Originally posted by: jliechty
Originally posted by: Stormgiant
I don't know if anyone has already done some tests with Barton's...

With a P4 you can't go wrong 😉

insteady of a 3.06, why not try a 2.8 or 2.66 and give it a bash in overclocking ?

a 2.8@3360 is lot faster than a 3.06@3450 and cheapper 😉
Click to expand...
P4s are good in SETI, if combined with Rambus or dual channel DDR, AFAIK, but in most other projects, they suck in comparison with Athlons. This difference is particularly noticable in RC5, ECCp/2, and Folding@Home (the only three I have seen many benches for). Aside from that, Athlons have a good price/performance ratio (consider my 1700+ TbredB that cost US$50 and almost makes it to 2GHz, 2400+).
Click to expand...

Actually the P4 only performs on par with the Athlon in SETI but in SoB and any other project sith SSE2 coding the P4 has a clear advantage.
 
The only comparison I can offer is of limited relevance, but....

I am on a mini-team and was able to see a first hand comparison of my AXP 2000+ (@1775) out-crunch a P4 2.4 (@ stock w/rambus)
The difference was small and I can only guess it must have been due to the FPU.

Anyway, both of these processors are a bit outdated, so glean whatever is meaningful and try to remember that a either a good Intel or AMD solution is going to do just fine. (exageration seems to be a theme around here lately when comparing hardware)
 
I couldn´t agree more with all of you...

AMD has one damm price/performance racio that intel can't reach for now, but like Baldy18 said, where
SSE2 and high FSB settings comes into play, P4 almost win every time.

AMD at stock speed beats almost every P4 on the streets but when you play with those P4.... hum, hum 😉
If you plan to overclock, those new P4 are really nice, if not, one fast XP+ is nice 🙂

I personaly don't have good experiences in setting up AMD for clients.
 
Stormgiant hit the nail on the head with what is important to me, price/performance ratio. Does it realy matter if Intel beats AMD by a little in performance if you have to pay twice as much for it? If you are going with Barton though the price difference isn't as dramatic since you are paying a high premium for them right now.
 
Actually Hyperthreading makes a huge difference in Seti. Barbary in his testing on a 2.2 Dual Xeon was getting 19WUs a day from a single box. It slowed the average time down but the number of results he was getting almost doubled.
 
Well for the moment this thread was just for some general info. i wasn't planning on upgrading until sledgehammer/intels' new cpu comes out...(running 1.2 ghz tbird , 786 mb pc-133 sdram..)

i've been an amd fan but in the end i will have to gouge price v. performance like always. if its less than ~10 % real world performance then is that extra premium really worth it?

but that's later in the year when i have some money! 😀:|
 
You must log in or register to reply here.

TRENDING THREADS

Back
Top