You need 290Fsb to be in the 230mhz mem range with 5:4 :(
288 = 230
personally I think 5:4 is a bad divider, i would have prefered a 6:5 option
Printable View
You need 290Fsb to be in the 230mhz mem range with 5:4 :(
288 = 230
personally I think 5:4 is a bad divider, i would have prefered a 6:5 option
You should be happy, I have to bench at 294 3:2 = 196 MHz memory :( :( :( :(
Auch, doesn't the Epox allow for some better benching then that ??Quote:
Originally posted by JCviggen
You should be happy, I have to bench at 294 3:2 = 196 MHz memory :( :( :( :(
Is't sacreficing some overall cpu power worth better memory?
not the EPoX, not the ABIT either.Quote:
Originally posted by Entuscan
Auch, doesn't the Epox allow for some better benching then that ??
Is't sacreficing some overall cpu power worth better memory?
highest I can bench 5:4 is around 284, thats 4 GHz. 4.1 GHz and 294 is a tad faster but not much... it would absolutely fly though at 5:4 but it seems to be impossible. Extra chipset voltage and cooling do absolutely nothing :(
Get me some R404A in the prometeia and I'll do 290 :DQuote:
Originally posted by Entuscan
You need 290Fsb to be in the 230mhz mem range with 5:4 :(
288 = 230
personally I think 5:4 is a bad divider, i would have prefered a 6:5 option
OPP
I agree...and disagree. If you ever deal with 'regular' customers, you know what I'm talking about ;)Quote:
Originally posted by JCviggen
on the other hand, it makes a lot of sense to maximize performance at stock speeds...... 90% of buyers will run stock, doing them a big favour. And from the looks of it, when overclocked the P4P is on par with the P4C .... also not bad
Just imagine...'I want the 875P, because it's faster'...'Yes, Mr. Customer, but this one miracle Asus 865PE is faster than all 875P's across the land!'...'I don't understand...'...'I don't either, Mr. Customer, it just is' ;)
In any event, OPP seems to have answered our question (though I'd like to know if you get the same funky results at stock as some of these web site are getting).
OPP's 3DMark2K1 and 2K3 scores at 277 with the Asus Springdale P4P800 are LOWER than his scores at 275 with his MSI 875P. This would seem to indicate that the P4P800 is 'reverting' back to normal Springdale-like results when overclocked.
Which answers my own question. Being that the P4P800 is same price as the IC7, the IC7 seems to be the better deal if overclocking.
Hey OPP...just for easy reference, for 5:4, multiply your FSB times 1.6 to get the mem speed. At 3:2, multiply times 1.33.
Quote:
Originally posted by Zroc
Hey OPP...just for easy reference, for 5:4, multiply your FSB times 1.6 to get the mem speed. At 3:2, multiply times 1.33.
Math is a b!tch isnt it ... ;)
I dont think they make any memory thÃ:banana:t fast yet
???Quote:
Originally posted by Zroc
Hey OPP...just for easy reference, for 5:4, multiply your FSB times 1.6 to get the mem speed. At 3:2, multiply times 1.33.
Why doesn't anybody just use the dividers and what they stand for
5:4 mem= fsb/1.25
3:2 mem= fsb/1.5
Funny im not having trouble with my IC-7 @ 292fsb 5:4 2,5,2,2 fast :D
MachII is due next week and i sold my normal Prommie so im out of action for a bit :(
Yes...apparently, OPP needs a comp faster than 4175 to perform those crushing computations ;)Quote:
Originally posted by JCviggen
Math is a b!tch isnt it ... ;)
Just ribbin' ya, OPP... :D
Quote:
Originally posted by toofast
Funny im not having trouble with my IC-7 @ 292fsb 5:4 2,5,2, fast :D
MachII is due next week and i sold my normal Prommie so im out of action for a bit :(
what CPU speed and CPU voltage...?
This epox board for example did 302 5:4 fastest settings with a 2.4C in a prom.
Jack it up to 4.1 GHz and you lose a whole lotta FSB :(
Well, you could do that...you'd need to multiply the result times 2.Quote:
Originally posted by Entuscan
???
Why doesn't anybody just use the dividers and what they stand for
5:4 mem= fsb/1.25
3:2 mem= fsb/1.5
It's just easier to go FSBx1.6 than FSB/1.25x2. Plus, a lot of motherboards list their ratios as 1.6 or 1.3.
Its a 2.8C JC @ 1.75v mem benchies will be about 65xx/65xx ish stoopid me forgot to do some :rolleyes:Quote:
Originally posted by JCviggen
what CPU speed and CPU voltage...?
This epox board for example did 302 5:4 fastest settings with a 2.4C in a prom.
Jack it up to 4.1 GHz and you lose a whole lotta FSB :(
Ah Zroc
Well know I get it, You're talking in DDR speeds/rating
I (and many others) like to talk about effectiev MHZ for memory
So then indeed both are correct
x1.6 or x1.3 = DDR speed
/1.25 or /1.5 = Mhz
Cheers :toast:Quote:
Originally posted by Entuscan
Ah Zroc
Well know I get it, You're talking in DDR speeds/rating
I (and many others) like to talk about effectiev MHZ for memory
So then indeed both are correct
x1.6 or x1.3 = DDR speed
/1.25 or /1.5 = Mhz
I suck at math so I just do it the long and easy to remember way. Take the FSB and divide by 5 then multiply by 4 if your ratio is 5:4.
Or take the fsb and divide by 3 then multiply by 2 to get your ram speed. Same goes with any fsb:mem multiplier.
Say fsb = 274 and your running 5:4 cpu to mem.
If fsb = 274/5 = 54.8 then * 4 = 219.2 mem speed.
Now that I look at it, it's kind of confusing and long heh but it works for any cpu to mem ratio and you do not have to remeber decimals.
I don't thin k the MSI is useing a 5:4 ratio. This could explain why this asus can't beat it even 60Mhz faster cpu speed. On the MSI when you set it to 333 it shows the mem speed in DDR, when you set it to like around 275 it would show 460 (230Mhz). If it was true mem speed it would explain why I needed so much mem voltage and why on this Asus I only need the measly 2.8V.
What would the ratio be with FSB at 275 and the mem at 230?
OPP
Hmmmm...
275:230
5:4.1818
But that margin is nominal. That difference falls into margin of variation in that test. Just 14 points of difference, I get 300 points variations if running 3dmark in a rowQuote:
Originally posted by OPPAINTER
As far as 3D03 the MSI seems to be stronger. I havn't done a clean install of Windows with this mobo yet so I may make up a few points.
MSI at 275 = 6158
Asus at 275 = 6144
OPP
For average user that gap in irrelevant, but of course for you, dudes, that margin can be everything :D
3D03 doesn't vary much at all, in fact I can get the exact same bench one after the other. The point spread being 10 points with the card not clocked is a lot. It's not like 2001 benching.
OPP
Actually, I think MSI's BIOS is just wrong (it says the same thing on the 865PE). It says the settings are 266, 333, and 400...but the 333 is really 320.Quote:
Originally posted by OPPAINTER
I don't thin k the MSI is useing a 5:4 ratio. This could explain why this asus can't beat it even 60Mhz faster cpu speed. On the MSI when you set it to 333 it shows the mem speed in DDR, when you set it to like around 275 it would show 460 (230Mhz). If it was true mem speed it would explain why I needed so much mem voltage and why on this Asus I only need the measly 2.8V.
What would the ratio be with FSB at 275 and the mem at 230?
OPP
The ratio if it really WAS 333 would be 6:5 (5 divided by 6, x2 = 1.66. So 200 x 1.66 = 333).
But the BIOS is just wrong. Even Intel calls 'the 333 setting', but it's really 320.
Looks like Abit has released a new IC7 BIOS..
Where about macci can it be found pls? :)
I can find a Beta IS7 one released 23/5 but no IC7 one
Cheers
:)
[Edit: Found it :DQuote:
Beta BIOS for IS7-G, IS7G_13.B01 (524,288 bytes)
New Fixes/Features
1. Supported IS7-G/IS7/IS7-E mainboard.
2. Updated CPU micro code (CPU: 0F29H, ID=11H).
3. Added CPU multiplier 8X for engineering sample CPU and maximum CPU
frequency 412 MHz.
4. Revised setup menu for on-chip Serial ATA.
5. Fixed on-chip SATA boot issue while another bootable hard drives exist
and attached to IDE-1 or IDE-2.
6. Updated memory reference code and ICH5 BIOS spec to v1.2.
7. Fixed PS/2 mouse malfunctioning problem while the USB cable of Media XP
attached to mainboard.
8. Fixed DOS 6.22 booting failure from FAT16 partition.
9. Fixed BIOS hang-up problem which is caused by specific CPU cooling fan
reporting error RPM to BIOS when CPU FanEQ 60% enabled.
10. Revised the following default settings in BIOS setup page.
- Bootable Add-in Device: PCI Slot Device
- CPU FanEQ Active Temperature: 70 C
- CPU Warning Temperature: 85 C
11. Added 80, 85, and 90 degrees C into "Shutdown Temperature".
12. Enabled "IDE Block Mode" and hided the option.
13. Revised ICH5R RAID BIOS display issue.
14. Revised power-on delay time from 4~5 seconds to 1~2 seconds after AC
power lost and resumed.
15. Added description for "Bootable Add-in Device".
Quote:
http://fae.abit.com.tw/eng/download/bios/dlbios.php?name=IC7&file=ic7/ic713.exe
BIOS release 1.3 for IC7/IC7-G v1.0
Release information:
[ENGLISH]
New Fixes/Features
1. Fixed DOS 6.22 booting failure from FAT16 partition.
2. Revised display mode of HTT CPU LOGO.
3. Fixed BIOS hang-up problem which is caused by specific CPU cooling fan
reporting error RPM to BIOS when CPU FanEQ 60% enabled.
4. Revised the following default settings in BIOS setup page.
- Bootable Add-in Device: PCI Slot Device
- CPU FanEQ Active Temperature: 70 C
- CPU Warning Temperature: 85 C
5. Enabled "IDE Block Mode" and hided the option.
6. Revised ICH5R RAID BIOS display issue.
7. Fixed COM2 device existing in Windows Device Manager problem.
8. Revised power-on delay time from 4~5 seconds to 1~2 seconds after AC power
lost and resumed.
9. BIOS Compile Date: 5/19/2003.
That would be exactly a 6:5 ratio. :toast:Quote:
Originally posted by OPPAINTER
when you set it to like around 275 it would show 460 (230Mhz).
What would the ratio be with FSB at 275 and the mem at 230?
OPP
Thats what it is then. This is the reason for topping out with the 333 at around 278 on the MSI, it matches my one to one ratio top out as far as mem speed.Quote:
Originally posted by MCWB
That would be exactly a 6:5 ratio. :toast:
OPP