Mhz Vs. Cas - Guess Whats Faster...

Printable View

Show 100 post(s) from this thread on one page

10-23-2006, 01:45 AM
sierra_bound

Update. 461Mhz is done. This is with 2.4v, not 3.0+ for memory voltage.

Can your Cellshocks do this at 2.4v?;)

http://img226.imageshack.us/img226/6...61x7333vm5.jpg
10-23-2006, 01:49 AM
hipro5

Quote:

Originally Posted by sierra_bound

Update. 461Mhz is done. This is with 2.4v, not 3.0+ for memory voltage.

Can your Cellshocks do this at 2.4v?;)

Unfortunetly no they can't do this any more after the 3.5Vdimm they had.....They work though but they are semi-died.....:D

Why do I have a fealing that we say the same thing?......1:1 with the most lower timmings we can use?...... :D
10-23-2006, 01:56 AM
sierra_bound

By the way, I'm not trying to get in a pissing match with you. I have a great deal of respect for your skills. But I think a distinction needs to be made between modded and unmodded boards. Micutzu did 490MHz @ 3-3-3 with 2.7v and the same TeamGroup memory. I'm just showing what can be done with an unmodded board.

The Pi times suck because I booted into Windows on the 1333MHz strap. I'm just trying to find the ceiling for this RAM. Then I will worry about tweaking.;)
10-23-2006, 03:15 AM
hipro5

No one is pissing off......We discuss here....:D :toast:

Now......

Can 460MHz with Cas 3-3-3-x come close to the Fatbodies at Cas 3-2-2-1....?.....NOT fully tweaked though (maxmem-etc).....Just a quick run that I did right now....PLUS that I have 2 x 512MB rams and not 2 x 1GB ones....PLUS that it was done on my retail unmoded ABIT IC7.....errrr.....AW9D-MAX I mean.....VMCH= 1.87Volts real.....;) :D

P.S. about 57MHz less MHz on rams..... :)

http://www.thelab.gr/images/Hipro5/C...14.46m_png.png
10-23-2006, 03:23 AM
hipro5

Quote:

Originally Posted by Kunaak

Super Pi 8M.
all settings done from bios.

I just chose one speed that I could use multiple fast mhz, and cas settings from. 7 x 367 worked out pretty good for me.

Hardware...

Asus P5B Deluxe. Bios 0711.
Mushkin PC 6400 2 gig kit, cas 4-4-3-10 set.
E6300 CPU.

from the looks of this, MHz is more important then cas alone, and 1:1 isn't better then other higher ratios...unless you can run 1:1 at over 1 ghz of course.

733 mhz at 1:1, cas 3-3-2-8 = 4M-10 Sec.

920 mhz at 4:5, cas 4-3-3-8 = 4M-3 Sec.

1100 mhz at 2:3, cas 5-4-4-8 = 3M-52 Sec.

EDIT: Saving alittle on the bandwidth, for member without high speed internet.
No Need to repost my own pics...

-Kunaak-

this is just a simple test, and not meant to be conclusive.

965 vs 975 chipsets......Not accurate coz the tests HAVE to be done on the same mobo/rams......That's why I've asked you to run your Fatbodies at 1:1 tight timings at 367MHz so as to compare yourshelf with your first bench.....;)

http://www.thelab.gr/images/Hipro5/C..._3.49m_png.png

I've said and I'm still saying that the companies are trying to "force" us believe that higher MHz Rams with looser timings are better, SINCE the TCCD time.....Back that time, they couldn't compete with the BH-6/5 chips and now they can't compete with the Fatbodies OR with the newer ram chips with cas 3-3-3-x......

BTW: at your third bench,the rams are at 550MHz and mine are ONLY at 367MHz thas 185MHz Ram frequency difference......975 vs 965 chipsets plus 1:1 ownage plus tight times ownage.....;)

EDIT: I have to add here that the "first" Cas, doesn't really matter a lot on Intel platforms....3 or 4 ....I'll test the 5 one so as to see what it does..... :)
10-23-2006, 03:37 AM
ozzimark

heh, from the looks of things, AM2 doesn't exist to you guys anymore :lol2:

i'll have to run a few benches myself, though i know 5-5-4 will end up being the fastest, since memory ratio means SH*T for K8 :toast:
10-23-2006, 03:43 AM
hipro5

Quote:

Originally Posted by ozzimark

heh, from the looks of things, AM2 doesn't exist to you guys anymore :lol2:

i'll have to run a few benches myself, though i know 5-5-4 will end up being the fastest, since memory ratio means SH*T for K8 :toast:

:D :toast:
10-23-2006, 04:16 AM
Kunaak

ok hipro, I will take this up further, cause yes... I do have the ram for it, and I am still curious.
10-23-2006, 05:00 AM
kiwi

Should be same cpu as well, allendale slower

Quote:

Originally Posted by ozzimark

heh, from the looks of things, AM2 doesn't exist to you guys anymore :lol2:

What's am2? :p:
10-23-2006, 06:28 AM
hipro5

Quote:

Originally Posted by Kunaak

ok hipro, I will take this up further, cause yes... I do have the ram for it, and I am still curious.

Yes please do.......out of couriosity.....:)
10-23-2006, 06:31 AM
hipro5

Quote:

Originally Posted by kiwi

Should be same cpu as well, allendale slower

Yes that's why I told Kunnak above to test by himshelf, to find out what will happen...;)
10-23-2006, 07:02 AM
MACMAC

Interesting thread.

Funny, I was going to make a post asking if there was a general consensus on what was better? Old Fatbody D9DQW/D9DQT or the newer D9GMH/D9GKX.

I have lots of 1GB Fatbody sticks, and some D9GMH, but no DDR2 system :slap:

Just trying to figure out which I should keep for future usage.
10-23-2006, 07:49 AM
fhpchris

Hipro dont tempt ozzy with cas 3-2-2, you know AM2 can only run 3-3-3 ;)

Thanks for all the effort that everyone has put into getting a solid comparison!
10-23-2006, 12:50 PM
Kunaak

3 Attachment(s)

ok, many benchmarks coming up, so please be patient while I post them all.

3DMark2001se Lobby.

http://www.xtremesystems.org/forums/...1&d=1161636469

http://www.xtremesystems.org/forums/...1&d=1161636469

http://www.xtremesystems.org/forums/...1&d=1161636469

Again, MHz Wins...
10-23-2006, 12:54 PM
Kunaak

3 Attachment(s)

Crystal Mark

http://www.xtremesystems.org/forums/...1&d=1161636720

http://www.xtremesystems.org/forums/...1&d=1161636720

http://www.xtremesystems.org/forums/...1&d=1161636720

Mhz Wins again...
10-23-2006, 01:00 PM
Kunaak

3 Attachment(s)

FEAR

Fear at 367, cas 3-3-2-8.

http://www.xtremesystems.org/forums/...1&d=1161636948

Fear at 460 cas 4-3-3-8.

http://www.xtremesystems.org/forums/...1&d=1161636948

Fear at 550 cas 5-4-4-8.

http://www.xtremesystems.org/forums/...1&d=1161636948

Quick Analysis.

going from 367 to 460, gets a higher Minimum frame rate, and a higher maximum frame rate but doesnt effect the average frame rate much.

going from 460 to 550, gets a higher average frame rate, and a higher maximum frame rate this time.

the difference between 367 and 550 is definatly enough to warrant using MHz more, then cas, when in a game.
10-23-2006, 01:15 PM
camouflage

Quote:

Originally Posted by NickS

Nice results, but welcome to last week.

:confused: Ever heard of being kind to other ppl.........?
10-23-2006, 01:24 PM
Kunaak

3 Attachment(s)

3dmark2003 Air Plane test... a personal favorite since this ones always been so sensitive to increases in ram or CPU Speed.

http://www.xtremesystems.org/forums/...1&d=1161638450

http://www.xtremesystems.org/forums/...1&d=1161638450

http://www.xtremesystems.org/forums/...1&d=1161638450

Again, MHz wins...
it looks pretty grim for cas, cause I am not finding a single case so far where Cas Alone will win a benchmark.

remember... this isn't Fatbody Vs. Regular D9...
thats a post for another person to make.
this is Mhz Vs Cas....
10-23-2006, 01:32 PM
fhpchris

Kunaak did you run your tests on p5B DLX?

If you didnt run memset and change the timings than that could be a possible solution, and I know memset and 965 has issues.
10-23-2006, 01:42 PM
bachus_anonym

Quote:

Originally Posted by hipro5

Get me a bench like this EVEN at 1300MHz DDR frequency....Not theory.....Fact.....;) :D

http://www.thelab.gr/images/Hipro5/C..._1_4_3_2_1.png

George,

Unfortunately, WinRAR is about one of the very, very few benchmarks where 2-3-3.0/4.0 shines that well and high frequency has a bit of a trouble catching up :)

Quote:

Originally Posted by hipro5

No one is pissing off......We discuss here....:D :toast:
Now......

Can 460MHz with Cas 3-3-3-x come close to the Fatbodies at Cas 3-2-2-1....?.....NOT fully tweaked though (maxmem-etc).....Just a quick run that I did right now....PLUS that I have 2 x 512MB rams and not 2 x 1GB ones....PLUS that it was done on my retail unmoded ABIT IC7.....errrr.....AW9D-MAX I mean.....VMCH= 1.87Volts real.....;) :D

P.S. about 57MHz less MHz on rams..... :) On P5B Deluxe, 2GB vs 1GB is amounts to 9-10s, so your 2-2-3.0 still wins by some. Don't get me wrongs, but I think Fatbody's days are over :fact:

http://www.thelab.gr/images/Hipro5/C...14.46m_png.png

I can tell you, that 358x9 DDR2-1074 5-4-4-4.0 on ASUS P5B Deluxe, pulls exactly the same time ;) No tweaks whatsoever, straight-up run on a default WinXP OS ;) Give or take 5-10 seconds for 1GB vs 2GB...
There is a point, where no matter how hard you will try to squeeze those 2's out of your RAM, high frequency will eventually win, especially with benchmarks where bandwidth is more important, like 32M... WinRAR is a whole different animal, though :)

EDIT: OK, 1sec off :lol:

http://www.clockmehigh.com/cooler/sc...5_4444_32M.png

Your run is still faster as I have a 9-10s advantage due to 2GB vs your 1GB, but don't get me wrong, but I think Fatbody's ultra-low latency days are close to over. Besides, most of those kits are 2x512MB, with just a short run of 2x1GB 8000UL. Now, I'd take D9GMH/GKX over Fatbody anyday :).
10-23-2006, 01:51 PM
nn_step

well Intel is less effected by latency than it is effected by Bandwidth.
AMD on the other hand is more effected by Latency than Bandwidth.
But honestly when more and more people get into overclocking DDR2 we will see that there are a hell of alot of tweaks needed to use it to its max
10-23-2006, 01:52 PM
sierra_bound

The 975 vs. 965 debate is getting very old. I have both. For RAM testing, the P5B Dlx is one of the best boards because you can hit memory speeds that unmodded 975X boards simply won't reach.
10-23-2006, 02:06 PM
Kunaak

Quote:

Originally Posted by fhpchris

Kunaak did you run your tests on p5B DLX?

If you didnt run memset and change the timings than that could be a possible solution, and I know memset and 965 has issues.

like I said in the first post...

all settings done from bios.
the only issue I am aware of, with memset and the P5B is that you can set cas 3 at any time, and it will do nothing... and you can do it at any MHz.

however, I dont understand what you mean by "Solution"...

there is no problem here to solve...
10-23-2006, 02:07 PM
Exedy

Quote:

Originally Posted by sierra_bound

The 975 vs. 965 debate is getting very old. I have both. For RAM testing, the P5B Dlx is one of the best boards because you can hit memory speeds that unmodded 975X boards simply won't reach.

and its just p5b dlx?

gigabyte ds3 and dq6 are in this situation?
10-23-2006, 02:09 PM
fhpchris

Quote:

Originally Posted by Kunaak

like I said in the first post...

all settings done from bios.
the only issue I am aware of, with memset and the P5B is that you can set cas 3 at any time, and it will do nothing... and you can do it at any MHz.

however, I dont understand what you mean by "Solution"...

there is no problem here to solve...

Read what Tony said again ;)

Quote:

Originally Posted by Tony

You guys are not taking advantage of sub timings, asus losen timings in different modes and ratio's so you have to set like for like to see a true comparison.

Once we have fully opened up bios files you will soon see the power of 1:1 with lower latency over upclocks with relaxed timings im sure. At this time we are being crippled by bios files...trust me ;)

Show 100 post(s) from this thread on one page