D975XBX Badaxe 1333fsb and 4:5

I see some folks say 4:5 on 1333fsb isn't working and think it's 1:1 but.........

I thought reason why Badaxe 4:5 can clock so much higher is that it is due to the less agressivenss of the MCH settings at 1333fsb to allow fsb to clock higher at least on badaxe... ?

Why else would at 1333fsb strap 1:1 is slower than 1333fsb strap 4:5 for me at same fsb and mem timings on cpu and ram ?

I am on 1333fsb strap and 2x512MB OCZ PC5400 Plat XTC max at around 423-426mhz at 2.2v on badaxe, when i tried 4:5 with 346fsb for 433mhz ram it wouldn't boot so i think 4:5 is indeed working.

i.e.

2x 512MB OCZ PC5400 Plat XTC maxes at 2.2v @423-426mhz 4-4-4-8/15

2x 512MB Corsair PC5400ULv1.5 Fatbody D9 Micron so far at 2.2v @506mhz 4-4-4-15 :eek: and still haven't tried higher yet.

Click image for full timings and mem screenshot
http://fileshosts.com/intel/Intel975..._single_tn.png

At 3649Mhz 9x405fsb and 4:5 506mhz 4-4-4-15 only 8 seconds slower than @3809Mhz (9x423fsb) and ram @423mhz 4-4-4-8 of 15min 05s!

5400ULv1.5 now at 3816Mhz 9x424fsb 4:5 with ram @530mhz 4-4-4-15 at 2.2v

http://fileshosts.com/intel/Intel975..._single_tn.png

Crazy 530mhz 4-4-4-15 at 2.2v which is 33 seconds faster than @3809Mhz (9x423fsb) and ram @423mhz 4-4-4-8 of 15min 05s :D

I think reason Badaxe clocks higher is MCH voltage can be raised to 1.725v whereas even stock Asus P5W DH only allows max MCH volts of 1.6v. To me badaxe 1333fsb 4:5 is NOT running 1:1 but indeed 4:5


Update: 1067fsb definitely faster

I wonder if how high fsb you can clock on 1067fsb is memory dependent ? i.e.

on 1067fsb with 2.2v vdimm 1304 bios

2x256MB Centon PC4300 Advance Micron D9 fatties = max 382fsb

2x512MB OCZ PC5400 Platinum XTC = max 378fsb

2x512MB Corsair PC5400ULv1.5 Micron D9 fatties = max 421fsb!

I've hit the limit of 1067fsb strap for 2x512MB PC5400ULv1.5 at 4-4-3-5 at 2.2v. The max is 421fsb. I guess you can subtract 20+ seconds if i had 2x1GB modules hehe.

http://fileshosts.com/intel/Intel975...3min53s_tn.png

good article, so it's not a whacked!!
Michael, you buy that yet?;)

Kev you try as well do same as well 1333fsb strap, same fsb but one test with 1:1 and 4:5 at same timings... bet the 4:5 is faster on badaxe :)

now 38 seconds faster heh

533mhz 4-4-4-15 at 2.2v

http://fileshosts.com/intel/Intel975..._single_tn.png

Quote:

---

Originally Posted by eva2000

Kev you try as well do same as well 1333fsb strap, same fsb but one test with 1:1 and 4:5 at same timings... bet the 4:5 is faster on badaxe :)

---

Yes that's what I talking about, but I am just afraid that michael doesn't think so ;)it definitely faster no doubt, unless he get a intel platform and run it . Thanks George!!!!

No probs Kev :)

Now the proof is in can anyone disprove your findings George :fact: :toast:

How about same benchmarks and 1066fsb strap? I don't buy it at all as it's common sense. Look at last ofcoolaler's threads with his PC2-8000 HZ... On Giigabyte mobo he manages DDR2-1000 5-4-4-3.0, which is fine and doable. He switches to BadAxe and bam - DDR2-1000 5-3-3-3.0 and probably even not 2.5v (he hasn't mentioned).

Do you guys have any idea how much voltage one would need for DDR2-1000 5-3-3-3.0? I'm not mentioning those "WR" :rolleyes: of DDR2-1100 8-2-2-3.0(!!!) or DDR2-1200 8-3-3-3.0. Switch to P2WD2-E Premium or P5W DH or AM2 boards and you will see how far in reality RAM clocks. I simply don't trust BadAxe overclocks, period! :) And don't buy a theory of loosening MCH timmings, either.

It literally NOT possible for FatBody to run DDR-1100 8-2-2-3.0 or new Micron at 1200 8-3-3-3.0 :fact: Also, it's totally off that suddenly Fatbody on BadAxe clocks 100Mhz higher than on other boards and new Micron D9XXX clocks on BadAxe waaay higher than on anything else. We're yet to see anyone hit DDR2-1000 8-3-3-3.0 on other Intel boards or AM2 but no, BadAxe has DDR-1200 already :lol: Someting is off with this board and 1333fsb starp...

But can you run at 1066fsb strap and 1333fsb strap George, please. Compare same FSB/CPU speed/ memory clocks results? The key is a strap, not neccesarly 4:5 ratio...

EDIT: Now, after I look at it one more time... Have you tried, tighter latenicies? 8-3-3-3.0 ? Maybe that's what important as well. Your DDR2-1080 8-4-4-4.0 is doable on other platforms as well so this one is not a problem. I have a problem with those huge, out of this world OCs on BadAxe at 8-2-2-3.0 or 8-3-3-3.0 :)

1067sb has a lower max fsb limit.. you won't be able to get fsb at 1067fsb anywhere near the the mem max....

1067fsb strap limit is around 378fsb on unmodded badaxe... 1333fsb strap has around 429-430fsb limit for me...

my badaxe has no mods and i haven't even shorted the oc debug pin so no vcore options available in bios heh

what's max MCH voltage on P5WD2-E ???

yes 1067fsb strap is more efficient clock for clock on badaxe.. look at my initial clocks with 1067fsb strap at http://www.xtremesystems.org/forums/...d.php?t=105003 - specific post http://www.xtremesystems.org/forums/...1&postcount=41

Tony/Bigtoe has already stated on intel platform 1067fsb strap is more efficient clock for clock than 1333fsb strap.

And you can't really compare mem performance on AM2 vs Intel, it's like saying because DFI NF4 can do TCCD >300mhz 2.5-4-3-7/2.5-3-3-7 then the TCCD can do that on say Gigabyte NF4 board hehe

Quote:

---

Originally Posted by eva2000

1067sb has a lower max fsb limit.. you won't be able to get fsb at 1067fsb anywhere near the the mem max....

1067fsb strap limit is around 378fsb on unmodded badaxe... 1333fsb strap has around 429-430fsb limit for me...

my badaxe has no mods and i haven't even shorted the oc debug pin so no vcore options available in bios heh

what's max MCH voltage on P5WD2-E ???

---

That's OK. I just want to see how much voltage you'd need for e.g. 375FSB and 4:5 ratio = ~DDR2-940 8-3-3-3.0 or 8-2-2-3.0. Also, isn't there 2:3 ratio ?

EDIT: Exactly my point... 1333fsb is what seems to be triggering those huge low-latency OCs.
I suspect AM2 is not much different and can be compared. People hit very high DDR2-1200+ 8-4-4-4.0, just waaaaaay lower @ 8-3-3-3.0 compared to BadAxe on 1333fsb. Besides, you don't need to look at AM2, just look at other boards for Intel ;)

well i only have 2.2v vdimm and even for PC5400UL i need 2.2v for 333mhz 3-3-2-8 LOL...

yes there's 800 mem option, haven't tried it yet as trying to figure out my max fsb first



Intel D975XBX Badaxe FSB and Memory clock calculator

http://i4memory.com/reviewimages/mot...Calculator.png

heh

If BadAxe is indeed such a monster for memory OCing then I don't want to see DDR2-1000+ 8-3-3-3.0 @ 1333fsb strap and 400+FSB @ 4:5 BUT DDR2-1000+ 8-3-3-3.0 @ 1066fsb strap and 333+FSB @ 2:3.

But it would be much better to see other board, that doesn't have 1333fsb in BIOS, to do it ;)

Heh i'll do some more investigating... just after i find the max for my conroe cpus hehe

Quote:

---

Originally Posted by eva2000

1067sb has a lower max fsb limit.. you won't be able to get fsb at 1067fsb anywhere near the the mem max....

1067fsb strap limit is around 378fsb on unmodded badaxe... 1333fsb strap has around 429-430fsb limit for me...

my badaxe has no mods and i haven't even shorted the oc debug pin so no vcore options available in bios heh

what's max MCH voltage on P5WD2-E ???

yes 1067fsb strap is more efficient clock for clock on badaxe.. look at my initial clocks with 1067fsb strap at http://www.xtremesystems.org/forums/...d.php?t=105003 - specific post http://www.xtremesystems.org/forums/...1&postcount=41

Tony/Bigtoe has already stated on intel platform 1067fsb strap is more efficient clock for clock than 1333fsb strap.

And you can't really compare mem performance on AM2 vs Intel, it's like saying because DFI NF4 can do TCCD >300mhz 2.5-4-3-7/2.5-3-3-7 then the TCCD can do that on say Gigabyte NF4 board hehe

---

Bingo, George, you push the right button.
that just give a true lesson on it;
well, for me, I rather buy the performance from intel, cuz Asus ...I just don't buy that either, no need to say other boards:)

Quote:

---

Originally Posted by eva2000

Kev you try as well do same as well 1333fsb strap, same fsb but one test with 1:1 and 4:5 at same timings... bet the 4:5 is faster on badaxe :)

now 38 seconds faster heh

533mhz 4-4-4-15 at 2.2v

---

What is the maximum that can you get with 1/1?

max for unmodded badaxe rev304 with no oc debug pin shorted is around 423-427fsb 1:1

http://fileshosts.com/intel/Intel975..._single_tn.png

Interesting stuff eva,

As I posted in the Intel section. I compared the P5W DH and Bad Axe with
identical hardware except the mobos of course and the Asus is far faster.

I was running the bad axe with 1333fsb and 4:5 ram.

Both at 3.5ghz with 1.4v and exact same ram and ram timings at 2.2v.

Asus:
http://www.jpop.com/gocchin/asus32m.png

Bad Axe:
http://www.jpop.com/gocchin/badaxe32m.png

That's even more dramatic than I expected. I took a look in Memset for a bit to try and tighten some secondary timings and noticed them both way tighter than I'm used to seeing. If you get time, you think you could post a shot of Memset on each of the boards?

try badaxe at 1067fsb and 350fsb 4:5 ;)

this was done at 2.2v vdimm 4-4-3-15 with 2x256MB Centon PC4300 Advance LOL with 1067fsb strap.. not sure how much a penalty 2x256mb would be heh

http://fileshosts.com/intel/Intel975..._single_tn.png

for conroe what's the diff between 2x256mb vs 2x512mb vs 2x1GB for 32m ?

also timings for tRAS differ it seems for your runs ?

I am going to do some more testing and see if what your saying is true. So far I have only used the 1333 then clockgen up from there. I noticed that with this mem I was getting instability at 4-4-4-8. So I started playing around with some timings and figured out that these (d9 fatbodies) like to run at 3-3-3-8... I have to find the ss but I ran a 32m at 440fsb 1:1 with those timings. Thanks for the info George I'll give this some testing later.

http://img455.imageshack.us/img455/4...3926ghz3mg.jpg

The differences are a little harder to spot with SuperPi 1M. 32m the differences are dramatic!

Oh I know I am looking for the ss now.. Its on one of these comps:) Looks like a few min difference there.

I hope now you guys are seeing that there is someting "disturbing" about memory performance and/or clocks on BadAxe @ 1333fsb strap, just as I suspect :p: :D This is why I think, forcing 1333fsb strap on BadAxe disqualifies it from serious and credible memory testing :fact:

well
just get a vmoded p5wd2-e vs vmoded D975xbx

i use 1066 vs 1333 on D975XBX there is no problem actually, 1066 is slower than 1333
but the point is we are still using D975XBX to do the test rams

if we want to find out there is problem with D975XBX, we need to use different mobo!

http://www.iamxtreme.net/cpu/e66/100_4335.JPG

http://www.iamxtreme.net/cpu/e66/100_4337.JPG

Here we go with 1067 all 3 screens together for comparo at 3.5ghz:

Asus P5W DH:
http://www.jpop.com/gocchin/asus32m.png

Bad Axe with 1333 bootstrap
http://www.jpop.com/gocchin/badaxe32m.png

Bad Axe with 1067 bootstrap
http://www.jpop.com/gocchin/badaxe1067.png

Any more questions? BadAxe @ 1333fsb vs 1067fsb and 1m16s difference :) This is a DRAMATIC change for 32M!!!

One other thing I noticed about using the 1333fsb is when booting the mobo powers off then on by iteself. I didnt have this problem before I started using the 1333fsb option... This only happens on a fresh boot not a reboot.. Does anyone else have this prob?

32m 1333+ i dont remember the %

OC only from bios

http://simon.webideal.ca/Bench/32m%203.96.JPG

Quote:

---

Originally Posted by njkid32

One other thing I noticed about using the 1333fsb is when booting the mobo powers off then on by iteself. I didnt have this problem before I started using the 1333fsb option... This only happens on a fresh boot not a reboot.. Does anyone else have this prob?

---

same thing happened here ....

Quote:

---

Originally Posted by Supertim0r

32m 1333+ i dont remember the %
OC only from bios http://simon.webideal.ca/Bench/32m%203.96.JPG

---

Another example here - your CPU is 450Mhz faster than gocchin's, you've lower latencies on RAM and yet only barely catching up with his P5W DH result...

BTW, in case some of you guys missed an interesting post from macci, regarding BadAxe 1066fsb vs 1333fsb:

Quote:

---

Originally Posted by macci

Someone mentioned Everest, latency and BadAxe...

Numbers I've seen (just an example speed):

1066strap + 343FSB + 343RAM, 3-2-2-8 = 52.7ns / 258.2FPS LH
1333strap + 343FSB + 343RAM, 3-2-2-8 = 69.8ns / 244.8FPS LH
1333strap + 343FSB + 428RAM, 3-2-2-8 = 71.4ns / 247.0FPS LH

---

Quote:

---

One other thing I noticed about using the 1333fsb is when booting the mobo powers off then on by iteself. I didnt have this problem before I started using the 1333fsb option... This only happens on a fresh boot not a reboot.. Does anyone else have this prob?

---

Its not a problem really - the board just needs to do that in order to switch to 1333 mode.

If there is any logic in the BadAxe BIOS it should give 1:1 when 1333FSB and 667RAM modes are selected (as they both mean 333MHz in reality). This is how it works with 1066 + 533. However if you select such a combination (1333+667) you'll see CPU-Z reading it as 4:5 RAM divider (FSB 333, RAM 417). Then again Everest doesnt show the clock info when 1333+667 is selected.

it could be possible that CPU-Z is reading the clock incorrectly w/ FSB1333 strap and 1333:667 is really 1:1 and not 4:5. This would mean that selecting 1333strap + 533RAM from the BIOS (which reads 1:1 in CPU-Z) would actually enable 5:4 RAM divider - this would put the clocks to 333FSB:266RAM

Of course someone needs to scope the clocks to verify this - perhaps hipro5 already did this?

Does the 800MHz mode ("2:3") work for you when you select 1333strap? The reason I'm asking is that I'm able to run the memory up to around 560-570 (CL4-4-4) using 1066strap and 2:3 divider (378FSB:567RAM) but it wont boot up at 1333strap:800MHz (333:500).

The performance w/ 1333strap is rather low (Everest mem bench):
400FSB "1:1" CL4-4-4 = READ:6737MB/s, COPY:5653MB/s, LATENCY:63.1ns
400FSB:500RAM "4:5" = READ:7151MB/s, COPY:6322MB/s, LATENCY:61.4ns
NB "X" @ 400 1:1 CL4 = READ:9242MB/s, COPY:6754MB/s, LATENCY:52.8ns

interesting macci, but it would not explain why my OCZ PC5400 Plat XTC which max at 423-426mhz wont boot at 1333fsb 4:5 at 346fsb, but my corsair pc5400ulv1.5 do boot and run because if it was 5:4 my OCZ ram would of worked fine ????

I have 3 bad axe. One of them won't run 400fsb 3-3-3-8 but my best run 445fsb 3-3-3-8 stable...weird

weird was this with same bios on all 3 badaxes ?

interesting findings so far, wonder if this does translate into 3dmarks as well. care to compare ?

:D time for me to post some results of BadAxe

my badaxe, cap/vdimm/vcore/tcore moded. bios=1304.

set 1067fsb, 50% host=400fsb w/o any problem and stable like a rock.
1067fsb, host=50%, ram:fsb=4:5 4-3-2-4, multi=11x, pi_32m=12min15sec
http://vic.expreview.com/attachment/1151504368.jpg

set 1333fsb, with a E6700, the mobo achieved 495fsb.
http://vic.expreview.com/attachment/...+ddr2_1237.jpg


then, some pi_1m results for you to compare:
1)1067fsb, host=50%, ram:fsb=1:1 3-2-2-4, multi=10x, pi_1m=12.781sec
http://www.xtremesystems.org/forums/...1&d=1151919492

2)compare to 1333fsb, host=30%(bios 1181) clkgen to 445fsb, ram:fsb=1:1 3-3-3-8, multi=9x, pi_1m=12.844sec
http://www.xtremesystems.org/forums/...1&d=1151920153

3)compare to ASUS P5WDH
364fsb, 11x, ram:fsb=1:1 3-3-3-8, pi_1m=12.828sec
http://www.xtremesystems.org/forums/...1&d=1151920597

Quote:

---

Originally Posted by eva2000

weird was this with same bios on all 3 badaxes ?

interesting findings so far, wonder if this does translate into 3dmarks as well. care to compare ?

---

Intel wins in CPU score of 3Dmark06, so far I've tested:

Quad 2.66G @ 320_10, cpu scroe=4999 with badaxe

Quad 2.66G @ 310_11, cpu scroe=4991 with p5w dh

and... for overclk ability/stability.

Intel wins both, compare with asus/msi/dfi

1.Highest fsb which can boot directly from bios.
2.same HIGHEST benchable clk with ASUS.
3.no vdroop, need 0.05v less to run dual_sp2004 with same cpu_clk compare to ASUS.

but isn`t it so Victor that a

cap/vdimm/vcore/tcore modded "bad axe"
seems to be only quite as fast as a unmodded asus P5W DH ? ....... :-)

and hmmmm....am i right , that your bad axe has selected handsoldered alu_solid_caps that replace most of the caps on the mobo ?

Quote:

---

Originally Posted by VictorWang

:D time for me to post some results of BadAxe

my badaxe, cap/vdimm/vcore/tcore moded. bios=1304.

set 1067fsb, 50% host=400fsb w/o any problem and stable like a rock.
1067fsb, host=50%, ram:fsb=4:5 4-3-2-4, multi=11x, pi_32m=12min15sec
http://vic.expreview.com/attachment/1151504368.jpg

---

Hi Victor. Here is an unmodded totally stock Asus P5W DH Deluxe.
4250mhz 10x425 Ram running 1:1 4-4-4-5
Even with 150mhz less and looser timings it's only slightly slower.

http://www.jpop.com/gocchin/1221.png

Quote:

---

Originally Posted by gocchin

Hi Victor. Here is an unmodded totally stock Asus P5W DH Deluxe.
4250mhz 10x425 Ram running 1:1 4-4-4-5
Even with 150mhz less and looser timings it's only slightly slower.

---

I saw ur thread yesterday and the result is damn good.:clap:
don't know if it's the higher fsb and higher ram_clk helps.

conroe is not sensitive to ram_timings 'cos of it has tooooo large L2_Cache :D

victor and gocchin what's the gains in 32m from going 2x1GB vs 2x512MB DDR2 at same timings/speed for ram ???

Victor I'm sure you will be under 12 minutes very soon! :)

eva, I think Kyosen-san had posted some times using SuperPi 8M with 2x512 and 2x1gb and the difference in 8M was around 5 seconds faster for the 2x1gb

Quote:

---

Originally Posted by eva2000

victor and gocchin what's the gains in 32m from going 2x1GB vs 2x512MB DDR2 at same timings/speed for ram ???

---

I've tested , 512m*2 vs 1G*2 at same settings pi_32m

1gb*2 leads 20sec way ahead :eek: :D

Quote:

---

Originally Posted by gocchin

Victor I'm sure you will be under 12 minutes very soon! :)

eva, I think Kyosen-san had posted some times using SuperPi 8M with 2x512 and 2x1gb and the difference in 8M was around 5 seconds faster for the 2x1gb

---

so that translates to >20s diff at 32M ??

I'm back on 1067fsb strap and with E6600 only having 9x multiplier had to do 9x389fsb 1:1 with ram at 389mhz 4-4-4-5 at 2.2v

seems with my 2x512MB PC5400ULv1.5 ram i can do 389fsb easy with 1067fsb strap straight boot from bios - stock unmodded badaxe rev304

http://fileshosts.com/intel/Intel975..._15m06s_tn.png

I loose 438-389 = 49mhz off memory clocks compared to your 10x350fsb 4:5 438mhz 4-4-4-5 and loose +20s off 32M with 2x512MB vs 2x1GB heh

Getting the hang of this board and cpu now, 1067fsb strap with 9x400fsb with 2x512MB PC5400ULv1.5 @400mhz 4-4-4-5 at 2.2v straight boot from bios with +50% burnin.

http://fileshosts.com/intel/Intel975...4min42s_tn.png

This is what it takes on Badaxe with 1304 bios for 2x 512MB modules on 1067fsb strap 1:1 cpu:mem to break 14min 28s single 32M. Booted straight from bios at 9x400fsb and used clockgen to raise fsb to 404.7fsb for 3646Mhz with ram @404mhz 4-4-3-5 at 2.2v

http://fileshosts.com/intel/Intel975...4min27s_tn.png


Crazy 1067fsb strap with 9x411fsb = 3703mhz with 2x512MB PC5400ULv1.5 ram @411mhz 4-4-3-5 at 2.2v totally unmodded badaxe rev304.

http://fileshosts.com/intel/Intel975...4min12s_tn.png

@9x415fsb 4-4-3-5 at 2.2v!

http://fileshosts.com/intel/Intel975...4min05s_tn.png

@9x418fsb 4-4-3-5 at 2.2v breaks 14min for 1st time :D

http://fileshosts.com/intel/Intel975...3min57s_tn.png

I wonder if how high fsb you can clock on 1067fsb is memory dependent ? i.e.

on 1067fsb with 2.2v vdimm 1304 bios

2x256MB Centon PC4300 Advance Micron D9 fatties = max 382fsb

2x512MB OCZ PC5400 Platinum XTC = max 378fsb

2x512MB Corsair PC5400ULv1.5 Micron D9 fatties = so far completed 32M upto 418fsb and doing 32M run @421fsb!

Quote:

---

Originally Posted by VictorWang

:D time for me to post some results of BadAxe

my badaxe, cap/vdimm/vcore/tcore moded. bios=1304.

set 1067fsb, 50% host=400fsb w/o any problem and stable like a rock.
1067fsb, host=50%, ram:fsb=4:5 4-3-2-4, multi=11x, pi_32m=12min15sec
http://vic.expreview.com/attachment/1151504368.jpg

---

talking about conroe mobos
found an interesting comparison :

_________________


coolalers 32 M run some time ago on a MSI 975 platinum


hmmm...300 MHz less and look @ the time...not bad at all :


http://img192.imageshack.us/img192/1523/coolaler6uv.jpg

Quote:

---

Originally Posted by mine

talking about conroe mobos
found an interesting comparison :

_________________


coolalers 32 M run some time ago on a MSI 975 platinum


hmmm...300 MHz less and look @ the time...not bad at all :


http://img192.imageshack.us/img192/1523/coolaler6uv.jpg

---

That looks to be just as fast as the Asus if not faster... I think we are all coming to the same conclusion here... the bad axe is slow....

gocchin try a 9x421fsb 1:1 4-4-3-5 run on asus i want to compare with badaxe stock unmodded with my custom 1304 1067strap bios


http://fileshosts.com/intel/Intel975...3min53s_tn.png

hehe

with clkgen, about 70% effiency in pi calculating clk to clk set from bios.
so, expecting gocchin and eva show us a 400*9 set from bios pi_32m
with ram 4:5 running at 500mhz 4-3-2-8.
that will be more easy for the comparision. :D:D

Quote:

---

Originally Posted by VictorWang

with clkgen, about 70% effiency in pi calculating clk to clk set from bios.
so, expecting gocchin and eva show us a 400*9 set from bios pi_32m
with ram 4:5 running at 500mhz 4-3-2-8.
that will be more easy for the comparision. :D:D

---

My corsair C3's arrive tomorrow... I have a much better chance at 500mhz 4-3-2-8 with them. :)

Quote:

---

Originally Posted by VictorWang

with clkgen, about 70% effiency in pi calculating clk to clk set from bios.
so, expecting gocchin and eva show us a 400*9 set from bios pi_32m
with ram 4:5 running at 500mhz 4-3-2-8.
that will be more easy for the comparision. :D:D

---

don't think i'd get my 5400UL to those speeds at 2.2v vdimm lol

ill see what my memory can really do tomorow :D
could someone remember where are the vdimm reading points ?

Quote:

---

Originally Posted by macci

NB "X" @ 400 1:1 CL4 = READ:9242MB/s, COPY:6754MB/s, LATENCY:52.8ns

---

How does "X" compare to the MSI 975 implementation?

We're ready for the boards anytime ;)

Quote:

---

Originally Posted by bachus_anonym

Any more questions? BadAxe @ 1333fsb vs 1067fsb and 1m16s difference :) This is a DRAMATIC change for 32M!!!

---

managed to push my E6600 + Badaxe rev304 a bit harder but still 2x 512MB PC5400ULv1.5 at 2.2v heh

http://fileshosts.com/intel/Intel975...4min37s_tn.png

Wow eva getting there! What did you do just tighten up timings and run 4:5?

i think what ram you use effects clocks and ultimately times


on 1067fsb with 2.2v vdimm 1304 bios

2x256MB Centon PC4300 Advance Micron D9 fatties = max 382fsb

2x512MB OCZ PC5400 Platinum XTC = max 378fsb

2x512MB Corsair PC5400ULv1.5 Micron D9 fatties = 421fsb

as you can see only corsair ram would of allowed me to do 9x389fsb in the first place..

for the 14m 37s run i loosened timings from 1:1 4-4-4-5 to 4:5 5-4-4-9 at 2.2v to run 4:5 for ram @486mhz.. cas 4 will reboot probably lack of vdimm heh

Went for my new personal best 32M time. I think this is one of the fastest times under 3.9ghz and 4.0ghz at 1:1 ??

http://fileshosts.com/intel/Intel975...9s609ms_tn.png

Heh, I guess we're getting off topic :lol: But I'm glad that there is consensus to 1333fsb being a memory performance killer and as macci said again, most likely running 1333fsb does clock-down memory frequency thru ratios. Just no software to show it :p: If the latest Everest 3.0 does not show RAM speed as he said, there there is definetely something off :fact:

nah don't think 1333fsb strap is running down clock divider otherwise my ocz p5400 plat xtc would of booted at >423fsb 4:5 upclock...

i had tested all day yesterday
with 1333, 533, 667, and 800

800 is not going to post for sure
667 is not running at 4:5, i think* it is runing 1:1 or 5:4 divider
533 is not running at 1:1 for sure but not sure the divider

I'm pretty sure that 1333 with 667 divider is 1:1. According to CPU-Z, same sticks of Teamgroup D9 2X1GB do:

- 420Mhz 3-3-2 @ 2.6V with 1066/667
- 525Mhz 3-3-2 @ 2.6V with 1333/667

If 4:5 is actually 1:1 at FSB1333, than in both cases the RAM is running at 420Mhz.

Rant mode on

Guys its nothing to do with clocking down the ram if you are in 1333 mode...its ALL ABOUT CHIPSET INTERNAL LATENCY

Do i have to follow you guys around and tell you like 7 times that its chipset related? come on....!!!!

Reduce the internal latency on the mem controller and the ram will work better. Find a way to force 1333 on the CPU strap but force 1066 on the NB strap and you will see ;) Asus obviously have found a way to do this.
This has been the same on all Intel chipsets since 845...you must see this nothing has changed.

Rant mode off...

Quote:

---

Originally Posted by Supertim0r

ill see what my memory can really do tomorow :D
could someone remember where are the vdimm reading points ?

---

nice and clean vmod :toast:

@ Super vdd measure half way down thread here:

http://www.xtremesystems.org/forums/...d.php?t=100125

Tony, but be realistic... Do you think, DDR2-1200 8-3-3-3.0 is even possible? Or DDR2-1150 8-2-2-3.0 ??? I don't believe that at all. Come on, no relaxing chipset latencies will do that, period! :fact: Everyone believing fatbody can clock to DDR2-1150 8-2-2-3.0 should be slapped :lol:

Look at the memory clocks P5W DH is showing and search for "WRs" on BadAxe, 1333fsb. You'll might change your mind.
:peace:

Quote:

---

Originally Posted by bachus_anonym

Tony, but be realistic... Do you think, DDR2-1200 8-3-3-3.0 is even possible? Or DDR2-1150 8-2-2-3.0 ??? I don't believe that at all. Come on, no relaxing chipset latencies will do that, period! :fact: Everyone believing fatbody can clock to DDR2-1150 8-2-2-3.0 should be slapped :lol:

Look at the memory clocks P5W DH is showing and search for "WRs" on BadAxe, 1333fsb. You'll might change your mind.
:peace:

---

get on skype I will explain so you can also....

Can anyone try Corsair XMS2 Xpert on the Bad Axe with 1333 strap ? Just to clear this up.

Quote:

---

Originally Posted by Dynasty

nice and clean vmod :toast:

@ Super vdd measure half way down thread here:

http://www.xtremesystems.org/forums/...d.php?t=100125

---

thanks ;)

http://www.xtremesystems.org/forums/...d.php?t=105703

i hope this helps

Quote:

---

Originally Posted by bachus_anonym

Tony, but be realistic... Do you think, DDR2-1200 8-3-3-3.0 is even possible? Or DDR2-1150 8-2-2-3.0 ??? I don't believe that at all. Come on, no relaxing chipset latencies will do that, period! :fact: Everyone believing fatbody can clock to DDR2-1150 8-2-2-3.0 should be slapped :lol:

Look at the memory clocks P5W DH is showing and search for "WRs" on BadAxe, 1333fsb. You'll might change your mind.
:peace:

---

A couple years ago I received some Platinum Revision 2 Enhanced Bandwidth Limited Edition DDR2 Modules from Alex Mey at OCZ - Some of the very first modules. These were equipped with Micron Fat Body IC's and I was able to push the memory to 7-2-2-3 @ DDR 960 on a DFI Board with modified bios....

So, I am personally rather sure that you could see DDR1150/1200 with more relaxed chipset latencies... :)

Just as I was talking to Tony I will say what I think here as well.

I'm not convinced that on Intel DDR2 can clock 30% better just from relaxing latencies between chipset and memory. There is so much chipset can do and so much RAM can do to allow for better memory OCs. In my mind, bringing AM2 example is a good one. Think about it...

AM2 is not restricted by latencies that on Intel mobos chipset brings when communicating with memory and CPU. Therefore, it might be safe to assume that DDR2 should clock on AM2 just as good as on Intel platform, If not even better. That assuming what Tony said, that chipset can make a WORLD of difference in memory OCing. I know it makes but I don't believe in such drastic changes as we see on BadAxe and 1333fsb strap, at least with monitoring software that we have on hand.

Now, with that said, something needs an explanation, don't you guys think? Why on AM2 new D9xxx does not seem to go higher than DDR2-980 5-3-3-3.0 @ over 2.7v so far but according to BadAxe results on 1333fsb strap, the same memory already "did" DDR2-1200 5-3-3-3.0. Then those same people (that tried BX) switched to ASUS P5W DH and we are yet to see DDR2-1000 5-3-3-3.0 on that mobo?

Also, I think that none of the software (CPU-Z, Everest and others) can properly detect 1333strap and therefore, like macci, hipro5 and few others are now saying - memory ratios might be different on 1333 than they are on 1066. I'm sorry, but I think believing that memory suddelny clocks a HEAVEN (not talking 10-20Mhz better but over 100MHz) better just on one freaky board will not convince me :nono:

Full respect to everyone :peace:

EDIT: PERFECT example! Same memory, same voltage but SPi32M @ SocketAM2 vs BadAxe @ 1333fsb strap 4:5

or Max clocks --> Shamino's 6400C3 @ 2.7v on SocketAM2 vs BadAxe @ 1333fsb strap 4:5 and 6400C3 @ 2.5v
vs Tyrou's BadAxe @ 1333fsb strap 4:5 and 6400C3 at 2.75v

Yeah, right! :rolleyes: Someone pleasssseee show me that on P5W DH or P5WD2-E Premium or new MSI or DFI Infinity.... :)

I agree bachus something isnt right. I know my mem can do quite well but I didnt expect 440fsb 1:1 3-3-3-8 at only 2.2v just doesnt add up to me. FUGGER ran this mem for awhile on his 955 and never got any results that I am... I hope that there will be a fix or something sometime.... Should I send you my setup so you can do some more testing maybe you can figure it out:)

@ Tony thanks for the explaining this.....

Its just the issue that you are trying to compare memory overclocking from one memory controller to another...simple fact is you can not.

Intel clocks ddr2 a whole lot differently than AM2 does, if you qualify DR2 on AM2 you can damn bet it will do a lot better on an Intel rig, same ram etc. Ram that will not do TRP at 3 on AM2 may do TRP at 2 on 955 or 975 at the same exact speed...this is a fact as we have seen it.Remember our VX UTT, modules that did 280 2-2-2 on A64 939 would not even do 220 on 875 Intel with the same vdimm so there are differences, its just DDR2 at this time clocks a lot better 2t on Intel over AMD.
DDR1 on A64 clocked a lot better than on Intel, but at this time it seems to be reversed for DDR2 although the AM2 memory controller is a lot more efficient even with higher latency on the modules.

Then you move to chipset latency, you hit the faster strap and the ram needs to drop to say 4-3-3- to clock, then you hit the slower strap and suddenly you can do 4-3-2 or even 4-2-2 at the same clocks...its weird i know but it has to do with the internal clock and latency of the memory controller on the NB.
Faster strap = lower strap IE running 800 instead of 1066, slower strap = higher strap so 1333 instead of 1066 etc

Intel NB need to be treated as a CPU. You have different strap ratio's that we need to call multipliers. As you clock up the fsb on your cpu you also clock up the fsb on the NB, eventaully it hits a limit and will not go any further. So you change the strap and it is able to start clocking again as in theory its clock speed has been reset to default. The only issue is it also sets the latency.
So you up the strap and lower the multi on the NB so you lower the NB clock.

Now PAT was just this as i said before, it increased the clock of the NB and this is why Asus had an ass slapping off Intel over the 865 chipset. Asus set the strap to 533 and changed the internal latency (available in 865 tweaker) and gained a huge performance boost, the only issue was no way would your BH5 clock as high unless you did all sorts of NB vmods etc so the board had a chance.... this is where Hipro5 had awesome success as he applied mods to the board that would allow the 533 strap on the IC7 and still have huge clocks with BH5...

So pick up here on the similarity, with the 533 or 667 strap on the IC7 you could run your ram at moderate overclocks but the performance was huge. You then set the 800 strap and the BH5 would do 30fsb more but performance was not as good...ask yourself why??? you downclocked the NB ;)

Now we are seeing some huge clocks with tight latency on Badaxe, the reason for this is 1333 strap on both the CPu and the NB...its probably set real slack latency...this is why you see 3-2-2-0 ddr1100 etc but performance will not be as good as 900 at 3-3-2- or 4-3-2 on the 1066 strap as the chipset is downclocked.

So some important rules to abide by here:
Don't compare intel to AMD memory overclocking...they are NOT the same.

keep the chipset strap low and the memory latency low and you gain a ton of speed, but you need an awesome NB, great memory and a good CPU to do it, and maybe a ton of mods ;)

One last thing, will someone please find and add the chipset latency tweak to the 975 tweaker, this will allow you to see what im partially talking about; While this will not increase the clock speed it will reduce the latency to the chipset.

To continue.

For the ultimate Intel 975 board, you would need an I/O chip between the CPU socket and the NB. The boot strap set by the BSEL pins on the CPU do not now go directly to the NB they go to the I/O chip. Bios Controls this strap and you are able to force any CPU strap you require.

Now the NB has to see the CPU strap but we have an I/O chip in the way, so we also control independantly thru bios the NB strap.This will allow us to force a NB overclock.

Normally the CPU strap is a direct link to the NB, some do have jumpers that allow you to force a different strap but the ultimate would be bios control over both.

I'm curious about something... Is there anyone willing to try how memory OCs differ at 800fsb vs 1066fsb strap on BadAxe?

Say, get your Conroe to start at stock 266FSB and then start looking for max memory OC at 800fsb strap. Then switch to 1066fsb strap and see if there is any difference in memory clocking and if there is then how big? How far Intel CPUs max out at 800fsb strap? 300MHz or so? That could be an indicator If 1333fsb in fact can make such a dramatic difference, if we are to believe in those huge OCs.

Quote:

---

Originally Posted by bachus_anonym

I'm curious about something... Is there anyone willing to try how memory OCs (pure frequency-wise, perhaps) differ at 800fsb vs 1066fsb strap on BadAxe?

Say, get your Conroe to start at stock 266FSB and then start looking for max memory OC at 800fsb strap. Then switch to 1066fsb strap and see if there is any difference in memory clocking and if there is then how big? Ho far Intel CPUs max out at 800fsb strap? 300MHz or so? That could be an indicator If 1333fsb in fact can make such a dramatic difference, if we are to believe in those huge OCs.

---

I had FCG do this already, the only issue was he did not document results...

well, with correct DDR2 1060 4-4-3-4, the 3.8G E66 with FSB1066 32M should be like this

E6600 with initial FSB1066 with GSKILL DDR2 800 2GBHZ
http://www.iamxtreme.net/cpu/e66/100_4344.JPG


compare with my old initial FSB1333..
http://www.iamxtreme.net/cpu/e66/100_4335.JPG

interesting 20s faster than my 3810mhz time of 13min 49s with 1067fsb strap and 2x512mb sticks heh

Quote:

---

Originally Posted by eva2000

interesting 20s faster than my 3810mhz time of 13min 49s with 1067fsb strap and 2x512mb sticks heh

---

1GBx2 is a little bit faster than 2x512 hehehehehe

Not bad.

Quote:

---

Originally Posted by Micutzu

I'm pretty sure that 1333 with 667 divider is 1:1. According to CPU-Z, same sticks of Teamgroup D9 2X1GB do:

- 420Mhz 3-3-2 @ 2.6V with 1066/667
- 525Mhz 3-3-2 @ 2.6V with 1333/667

If 4:5 is actually 1:1 at FSB1333, than in both cases the RAM is running at 420Mhz.

---

That makes a lot of sense.

Until someone has actually measured the Memory clock in 1333:667 mode (which quite logically should be 1:1) there is not much point in this discussion.

Quote:

---

Originally Posted by Micutzu

I'm pretty sure that 1333 with 667 divider is 1:1. According to CPU-Z, same sticks of Teamgroup D9 2X1GB do:

- 420Mhz 3-3-2 @ 2.6V with 1066/667
- 525Mhz 3-3-2 @ 2.6V with 1333/667

If 4:5 is actually 1:1 at FSB1333, than in both cases the RAM is running at 420Mhz.

---

I can confirm that
I think 1066/533 is 1:1 and 1333/667 is 1:1 too
But unfortunally CHIPSET INTERNAL LATENCY is much worse @ 1333 than @ 1066 and performace is still crappy.
I need 3522MHz (1333/667) to have similar performace like 3388MHz (1066/533) - mems timings identical /meansured in PI 1M&16M/

I've posted this in another thread also:

E6600 @ 4.3Ghz, 2X 7900GTX @ 850/1070
FSB 1333 + RAM @ 597 3-4-3-4 vs FSB1066 + RAM @ 477 3-4-3-4 (RAM clocks reported by CPU-Z)

http://www.xtremesystems.org/forums/...1&d=1152697345


If what i've said before applies and the RAM is actually running at the same clocks in both cases, there's still something else involved since LH gets a consistent boost with FSB1066. On a side note, i needed higher voltages on MCH/RAM to run at FSB1066 setting.

Hey guys, nice thread, I am having problems with the 1333FSB strap on my latest bad axe (last one died)
It does the power off thing but seems to keep booting at 1066 whatever settings I try. Can anyone advise the secret to getting it to work on is this just becuase of the latest BIOS version 1304?
I can get 400FSB x8 prime stable, with +50% limit on 1066 strap using this unmodded board and 1304, 400x9 wont even work with 1.45 vcore! 266x13 is 90% stable at stock vcore also. This is an X6800 (es) Step 5 B1.

Stumbled on to this last night with IDCC.

Look at this below. I'm running the 1333 strap + 9% = 363fsb with mem at 667 which should be 4:5 ratio or 454MHz on the mem.

CPUz says it's 454Mhz.
IDCC says mem is DDR2 727 or 363Mhz which is 1:1 ratio

Which one is right :)

http://img221.imageshack.us/img221/9...63fsb45mk3.jpg

Bingo Blue078! :up:

It looks like you might have just delivered THE confirmation we all needed to prove what some of us suspected (DDR2-667 ratio and 1333fsb strap = 1:1, not 4:5)! Pretty funny, that nobody else with BadAxe has actually tried this soft before :)

installing it right now...

edit : yes it show the "REAL" memory speed, nice find Blue078

Still at 1333 strap but this time chose the 533 mem option.

Now which is right ?

http://img101.imageshack.us/img101/2...b533memds9.jpg

In both cases 533 and 667 sandra show the same as IDCC.

http://img60.imageshack.us/img60/538...1333533lw2.jpg

http://img158.imageshack.us/img158/8...3331145oa0.jpg

Quote:

---

Originally Posted by Blue078

Still at 1333 strap but this time chose the 533 mem option.

Now which is right ?

http://img101.imageshack.us/img101/2...b533memds9.jpg

---

IDDC again ;) 5:4 instead of 1:1. The suspicion some of us had was that at 1333FSB strap (333Mhz) memory ratios would look like this (I'm not sure If DDR2-800 can be selected on BadAxe as 975X does not officially support DDR2-800):

DDR2-533 --> 5:4 (266MHz)
DDR2-667 --> 1:1 (333MHz)
DDR2-800 --> 4:5 (400MHz)

I wasn't trying to prove anyone right or wrong.

I've lost 60fps in CS:S's VST moving from 3800-X2/SLI-D/1900XTX to E6700/BadAxe/1900XTX. and have been trying every BIOS combo trying to sort out the loss. Then I stumbled on to this. My next step is to drop back to the 1066 strap and see if I pick up the lost FPS.

Quote:

---

Originally Posted by Blue078

I wasn't trying to prove anyone right or wrong.

I've lost 60fps in CS:S's VST moving from 3800-X2/SLI-D/1900XTX to E6700/BadAxe/1900XTX. and have been trying every BIOS combo trying to sort out the loss. Then I stumbled on to this. My next step is to drop back to the 1066 strap and see if I pick up the lost FPS.

---

No worries... You have actually made all of us a HUGE favor by showing what you have found out! We have had quite a debate here regarding actual memory frequency and ratios when forcing 1333 strap from BadAxe's BIOS.
Thanks man! :toast:

OK their is a 800 mem option showing while set to 1333 strap. Would like to see what ratio it comes up with but cant get the board to POST even with 5-5-5-15 timings if I choose the 800 option :)

I'm running the Corsair 8500C5's and haven't had any trouble running them in the 1066 strap at 504MHz 5-5-4-14 so not so sure why the no POST..

Sorry someone with better mem will have to find this one out :toast:

Edited goofed on the 6-6 changed to 5-5

Nice find Blue078 ;)
Bad Axe doesn't boot with the FSB1333 + DDR800 setting btw.

...this isn't the first time that CPUz has 'read' the wrong settings.

Not to bash the program (we'd be practically no where without it, let's be honest), but its methods of calculating speeds are JUST THAT....calculations based on apparent set settings. *NOT* readings of actual speeds.

Now, it's not all CPUz's fault (in fact, it's mostly the mobo makers') that this causes problems: motherboard makers don't necessarily use a standard for clock and ratio settings....if a setting name directly corresponded to what it should do (like the "180" mem ratio on K8, especially when used with the 11x multi to get "/13" mem divider), then CPUz would be right....but since a BIOS setting is only a pretty face, lots of ugly stuff can happen and CPUz won't detect it until someone comes along and shows the TRUE speeds, so that the program can be selectively reprogrammed :)

That's where we come in....great find Blue078 :toast:

Quote:

---

Originally Posted by eva2000

I see some folks say 4:5 on 1333fsb isn't working and think it's 1:1 but.........

I thought reason why Badaxe 4:5 can clock so much higher is that it is due to the less agressivenss of the MCH settings at 1333fsb to allow fsb to clock higher at least on badaxe... ?

Why else would at 1333fsb strap 1:1 is slower than 1333fsb strap 4:5 for me at same fsb and mem timings on cpu and ram ?

---

Hi,
divisor 667 of strap 1333 = 7:8 ??
divisor 533 of strap 1333 = 8:7 ??
is it possible??


sorry but my english is very poor :slapass: :slapass:

thanks

1067fsb strap 533 = 1:1

Radiical chipset waterblock with fresh layer of Ceramique thermal paste on Intel Bad axe MCH/NB core (no volt mods still). Best of all I can now do 488mhz 4-4-3-5 at 2.2v with very tight tRFC value of 25!

http://fileshosts.com/intel/Intel975...9s391ms_tn.png

With better MCH cooling seems my memory is flying even at 2.2v vdimm :D

Here's what clocks it took to break 15mins :)

http://fileshosts.com/intel/Intel975...8s781ms_tn.png

I can't believe you all though 667 in 1333 strap was 4:5...come on..LOL

in 1066 strap 533 = 1:1
in 1333 strap 667= 1:1

I publish an article trying to explain this crap showing how the strap changes the memory ratio's and you still debate on and on on an on an on...

So I will cut thru the crap and tell you all what you need the motherboard to do.

You need the bios to force the 1333 strap on the clockgenerator, this will allow it to scale easy over 370fsb....normally the max will be 100fsb over this.

You need the bios to NOT set the 1333 strap on the NB but to set the 1066 strap on the NB instead, you will still have the awesome speed the ram can do but you may have to run 1:1 instead of 4:5.

The sooner RD600 comes out the better, this strap 5h1t is killing me now..LOL

Quote:

---

Originally Posted by Tony

The sooner RD600 comes out the better, this strap 5h1t is killing me now..LOL

---

hehe i can just imagine hehe

Quote:

---

Originally Posted by Tony

I can't believe you all though 667 in 1333 strap was 4:5...come on..LOL

---

Hahaha, that's what the whole fuss was about and some of us wanted to prove in wrong, Tony! :) It was that, CPU-Z made man, many people believe (starting with early adopters of BadAxe), that 1333fsb strap forced from BIOS and ratio DDR2-667 selected at the same time, yielded 4:5. That's what Memory Tab was showing, indicating huge (and impossible) CAS3 overclocks and some of us were trying to prove that it was wrong - correct info there should be 1:1. I think you have misunderstood the whole thing :)
But I'm glad we're finally on the same page now :toast:

Quote:

---

Originally Posted by guess2098

1GBx2 is a little bit faster than 2x512 hehehehehe

---

NOT just little bit, it's dang A LOT:D

Quote:

---

Originally Posted by bachus_anonym

Hahaha, that's what the whole fuss was about and some of us wanted to prove in wrong, Tony! :) It was that, CPU-Z made man, many people believe (starting with early adopters of BadAxe), that 1333fsb strap forced from BIOS and ratio DDR2-667 selected at the same time, yielded 4:5. That's what Memory Tab was showing, indicating huge (and impossible) CAS3 overclocks and some of us were trying to prove that it was wrong - correct info there should be 1:1. I think you have misunderstood the whole thing :)
But I'm glad we're finally on the same page now :toast:

---

I may have been a little off base but you will notice cas3 clocks will be higher using 1333 strap, TRCD at 3 also will be possible where using 1066 will will have to losen to 4.

If you check the guide I wrote it shows what ratio's are available per strap, if I get time I will work out and add the special ram ratio's like you see on the asus etc and show how they get them.

Side note, if you get a 1333fsb CPU and you use one of the special ram ratio's you are actually forcing the NB to a different strap, so this may yield extra speed.