Testing / comparing : Intel D975XBX2 / Asus P5B DX ***56K WARNING***

Quote:

---

Originally Posted by ChaosMinionX

http://i11.photobucket.com/albums/a1...ionX/intel.jpg




Well be aware that the 3 ports on the bottom left hand side of the above picture are USB2 and 1394. However there is no way to delineate between the 1394 and USB2 because they are both freakin blue (GG Intel you retards) so the 2 directly next to eachother next to front panel hookups are USB2, the one at the bottom edge of the board by itself is 1394.

I recently have killed 4 ipod nano 8gb because my top usb input was hooked up to a freakin poorly marked 1394. Just a warning to anyone else that may have made the same mistake. Intel could have at least made it red or some other color like the other 90% of mobos.

---

The hell? My USB headers are black, and so is the Firewire I think, although it may be red or something. And my SATA ports aren't light blue, either. I thought color schemes were the same on all like-revision boards. :confused:

Quote:

---

Originally Posted by ChaosMinionX

Well be aware that the 3 ports on the bottom left hand side of the above picture are USB2 and 1394. However there is no way to delineate between the 1394 and USB2 because they are both freakin blue (GG Intel you retards) so the 2 directly next to eachother next to front panel hookups are USB2, the one at the bottom edge of the board by itself is 1394.

I recently have killed 4 ipod nano 8gb because my top usb input was hooked up to a freakin poorly marked 1394. Just a warning to anyone else that may have made the same mistake. Intel could have at least made it red or some other color like the other 90% of mobos.

---

Not their fault you can't read. That's your own error man. And don't be so :banana::banana::banana::banana:y either. If you had looked in the manual, you would have seen how to hook it up, and what it was.

Quote:

---

Originally Posted by wa2000

Change the "default" to 333 and change the 533 to 667. This will switch the nb to 1333 strap and keep the fsb:memory at 1:1.

---

i changed the default to 333

but at 450fsb ... 667 makes my ram run at 4:5 or 5:4 wutever it is .. which makes it run at 1160 when im at 450 .. so i kept it at 533 which runs the ram at 1:1 which would be ddr2 900

Quote:

---

Originally Posted by Reinvented

Not their fault you can't read. That's your own error man. And don't be so :banana::banana::banana::banana:y either. If you had looked in the manual, you would have seen how to hook it up, and what it was.

---

QFT. 15 seconds in the manual prevented that from happening to me. There's more important matters to :banana::banana::banana::banana::banana: about with this board, ones that don't have answers in the manual.

Quote:

---

Originally Posted by jetjaguar

i changed the default to 333

but at 450fsb ... 667 makes my ram run at 4:5 or 5:4 wutever it is .. which makes it run at 1160 when im at 450 .. so i kept it at 533 which runs the ram at 1:1 which would be ddr2 900

---

Try to use IDCC to check the memory speed. Because I remember I read somewhere in this thread that CPUZ won't report the correct fsb:mem ratio when the board is in 1333 strap.

Quote:

---

Originally Posted by Reinvented

Not their fault you can't read. That's your own error man. And don't be so :banana::banana::banana::banana:y either. If you had looked in the manual, you would have seen how to hook it up, and what it was.

---

First of all you shouldnt really have to look at the manual since USB and 1394 are pretty straight foward, and hard to mess up but when they mark them the same and the 1394 is inconspicuously placed it can be a little easy to do. They mobos dont change that much that you need to read instructions for it everytime you buy a new one, second 1394 usually by a sensible manufacturer a different color than the USB just for quick hook ups and removing this type of situation. It wasnt a rant, I could careless about the Ipods I can buy more, your post really had no point to it other than you being a jackoff.

And your use of whatever you said "And don't be so y either." try using something more intelligent that doesnt get filtered. I was merely point it out to others that may have the same thing so they dont kill stuff.

I know what im doing, and it just happens to be poor mobo layout I can name a dozen or better mobos that use 2 diff colors. And im sure others would agree its a little bit moronic on intels part to not take that into consideration, considering who might buy their mobo.

And dont be a smartass in your post it doesnt make you look very good just making posts like that, you really didnt contribute to anything at all. Just like to push your nonsense in a post and think your above making an error like that or something. I can read, It should be better marked than being all the same color and rather close to eachother. Im sure im not the only one that does it, but thanks you for your useless post Reinvented youve upped your count by 1.

Quote:

---

Originally Posted by wa2000

Try to use IDCC to check the memory speed. Because I remember I read somewhere in this thread that CPUZ won't report the correct fsb:mem ratio when the board is in 1333 strap.

---

IDCC ?

damn 3.6 died on orthos at 1hr 45 min sucks

Quote:

---

Originally Posted by ChaosMinionX

First of all you shouldnt really have to look at the manual since USB and 1394 are pretty straight foward, and hard to mess up but when they mark them the same and the 1394 is inconspicuously placed it can be a little easy to do. They mobos dont change that much that you need to read instructions for it everytime you buy a new one, second 1394 usually by a sensible manufacturer a different color than the USB just for quick hook ups and removing this type of situation. It wasnt a rant, I could careless about the Ipods I can buy more, your post really had no point to it other than you being a jackoff.

And your use of whatever you said "And don't be so y either." try using something more intelligent that doesnt get filtered. I was merely point it out to others that may have the same thing so they dont kill stuff.

I know what im doing, and it just happens to be poor mobo layout I can name a dozen or better mobos that use 2 diff colors. And im sure others would agree its a little bit moronic on intels part to not take that into consideration, considering who might buy their mobo.

And dont be a smartass in your post it doesnt make you look very good just making posts like that, you really didnt contribute to anything at all. Just like to push your nonsense in a post and think your above making an error like that or something. I can read, It should be better marked than being all the same color and rather close to eachother. Im sure im not the only one that does it, but thanks you for your useless post Reinvented youve upped your count by 1.

---

Yeah bud, you must know what your doing since you killed four ipods. And sir, you haven't contributed much yourself. And I am above errors like that, since I do read the manual and take my time with it. It's also clearly marked on the board too, but you wouldn't have noticed it because you "know what you are doing". GG retard.

Also - That's a picture of an XBX...not an XBX2. They are still labeled on both revisions.. Maybe you should get a pair of READING glasses.

Quote:

---

Originally Posted by jetjaguar

IDCC ?

damn 3.6 died on orthos at 1hr 45 min sucks

---

Intel Desktop Control Center
Download it here: http://www.intel.com/design/motherbd.../dcc/index.htm

Quote:

---

Originally Posted by ChaosMinionX

First of all you shouldnt really have to look at the manual since USB and 1394 are pretty straight foward, and hard to mess up but when they mark them the same and the 1394 is inconspicuously placed it can be a little easy to do. They mobos dont change that much that you need to read instructions for it everytime you buy a new one, second 1394 usually by a sensible manufacturer a different color than the USB just for quick hook ups and removing this type of situation. It wasnt a rant, I could careless about the Ipods I can buy more, your post really had no point to it other than you being a jackoff.

And your use of whatever you said "And don't be so y either." try using something more intelligent that doesnt get filtered. I was merely point it out to others that may have the same thing so they dont kill stuff.

I know what im doing, and it just happens to be poor mobo layout I can name a dozen or better mobos that use 2 diff colors. And im sure others would agree its a little bit moronic on intels part to not take that into consideration, considering who might buy their mobo.

And dont be a smartass in your post it doesnt make you look very good just making posts like that, you really didnt contribute to anything at all. Just like to push your nonsense in a post and think your above making an error like that or something. I can read, It should be better marked than being all the same color and rather close to eachother. Im sure im not the only one that does it, but thanks you for your useless post Reinvented youve upped your count by 1.

---

If you can afford that many ipods, why not just get a motherboard that doesn't have a "poor layout" instead of crying to us about how you're to good for reading.

Thread crap - check!

Quote:

---

Originally Posted by ChaosMinionX

Thats odd, have you tried changing your vmch a bit since your FSB is higher to give the memory lil extra boost? i had to bring my vmch up to make dominatorc3 6400's run at advertised with 0 memtest 86+ errors. But I wouldnt be suprised, XBX2 seems to have killed my Dominators that are currently in the process of being replaced. They ran fine on my P5B, then I put em on XBX2 and they died so not really sure, Ive also checked the dimms with a MM and it read +/- .03-05v as set in bios. Try the Vmch and let me know how it turns out be interested to know.

---

Thanx for help chaos. Well i brought up vmch and no change. The alpha titanium is dead..to bad it was the best out of all of them. The other sticks run at 400fsb\533\266. But they will not run at 333fsb\533\200. Oh well i guess its rma time. Wierd cus alpha sticks were never higher then 2.3v for any length of time. And they were the only ones that could run on the 800 strap.

afireinside / Reinvented / ChaosMinionX

If you just don't read each other's posts and don't respond to trolling, it will all disappear and we can get back to purposeful discussion.

is it worth going from 2507 to 2333 ??

Quote:

---

Originally Posted by jetjaguar

is it worth going from 2507 to 2333 ??

---

I saw no problems. And it's been rock solid for me. Besides, you can always try it out, and if you don't like it, you can flash back.

Quote:

---

Originally Posted by jetjaguar

is it worth going from 2507 to 2333 ??

---

In my case nope.

2507 is the most stable for me. 2330 and 2333 were not stable for me.

2507 and 2496 seam to be comparable, but 2507 seams to be more stable after several days of being on running folding or prime. 2495 would give me a random reboot after being on a for about 12 to 15 hours 2507 has fixed that for me.

After all my issues and stuff, I'm fully 100% rock solid several days of orthos stable at 413Mhz FSB on a E6600 which runs about 3.713Ghz. 4 GB of Crucial Ram running 1:1 which is DDR826 @ 4,4,4,10 with 2.24 Volts. I've been running Dual Prime95 for a week straight and no more degradation.

BIOS settings I've found to matter are:
C1E - Disabled
Watchdog - Disabled ( see explanation below why this WONT WORK!!)
Debug - Disabled
PCI Bus Latency - 64 (seams to help with USB devices and HDTV card lockups)
PCI express burn in - 101.x (seams to have helped lockups during games over default, anything higher than 101 gives me lockups)
Vmch - 1.7v actual from Vmod (I think I just have a weak and wussy 925X)
Vfsb - 1.425v (Helped with stability although I'm still testing this)
VT - enable (computer seams more stable after enabling even though it should have NOTHING to do with stability... maybe its my wussy chipset)
Vmem - 2.24 (usual stuff here nothing new)
Vcore - 1.45 (1.400 actual measured usual stuff here)

For the stupid "watchdog" setting on the XBX2, it will not work correctly on anything needing slightly hot voltage for the CPU WHY?? Well the XBX2, like most motherboards, has an initial powerup Vcore voltage that is per Intel specification of 1.1xx volts. Well if your CPU needs 1.45 volts to be stable at a certain overclock, then 1.1xx volts aint gonna get it stable enough to post correctly so the CPU clock will not be clocking and the watchdog will not get it's clock signal so the watchdog timer (thats on any commercially available computer circuit made by mankind) will pulse the reset line, by now the motherboard has given the CPU the higher VID voltage or cmos programmed Vcore voltage and now posts.

The intel "watchdog" is basically monitoring whether the watchdog is firing during post, if it fires, then it assumes that you have a poor overclock and gives you that stupid error. Watchdog timers are on any and every motherboard ever made by mankind and any and every embedded CPU, microcontroller, PIC, etc. It is nothing more than a system that watches for a clock signal and if it stops for a time length, it fires a reset signal in hopes of restarting the clock signal.

Using the watchdog "feature" on the XBX2 will limit your overclocks to a low enough setting that still enables the CPU to initially clock correctly at 1.1xx volts thus avoiding the watchdog from firing.

On other motherboards (p5b comes to mind) some people have problems with their system not posting until they hit reset. In this case, you are actually acting as a "watchdog timer". You physically see that there is no post which means there is no clock so you pulse the reset line and then things work because by the time you push the reset line, the Vcore is up where it needs to be in order to post.

Basically on the XBX2, the watchdog is a feature when it is disabled in the BIOS because the disabled setting allows the watchdog to reset the computer and give it another shot at posting when Vcore is higher, which is how a watchdog circuit should work in the first place. This concept unfortunately is not something new and has been around since before I was a fetus. Intel's watchdog setting basically tells you that the watchdog timer has fired a reset signal during post which means that the CPU isnt stable at the specified 1.1xx volts intial startup voltage. I basically see it as a way for Intel to limit Overclocking for a huge majority of casual overclockers / enthusiasts, thus making them want to upgrade to a better CPU in the future.

Quote:

---

Originally Posted by Reinvented

Yeah bud, you must know what your doing since you killed four ipods. And sir, you haven't contributed much yourself. And I am above errors like that, since I do read the manual and take my time with it. It's also clearly marked on the board too, but you wouldn't have noticed it because you "know what you are doing". GG retard.

Also - That's a picture of an XBX...not an XBX2. They are still labeled on both revisions.. Maybe you should get a pair of READING glasses.

---

OT - But maybe you should stop being arrogant. And get reading glasses and find out that im not whining about the board, im just stating a flaw that could possibly cause someone else to ruin ipods or any other usb device. I dont see how my posts havent really helped, but ok your brain works on a higher level than mine must be mr hawking.


Ok so yes im not whining about the board, and the fact that I can get more Ipods doesnt really say anything about money, more so about a good return policy and within 30days of buying it. but yeah lets get back on topic and stop flaming. :slapass:


Also - XBX and XBX2 look identical with a few minor changes, and the usb and 1394 ports on my board are in fact blue, not black. So I just posted the quickest pic I could find as I was on lunch when I made the post.

ChaosMinionX: Why not just let this die...

I did. That was the last post concerning that :rolleyes:

Since both sanda and everest dont work with overclocked xbx2, which memory benchmark do u suggest?

Quote:

---

Originally Posted by cyber_terror

Since both sanda and everest dont work with overclocked xbx2, which memory benchmark do u suggest?

---

Read post #1613

Quote:

---

Originally Posted by ChaosMinionX

OT - But maybe you should stop being arrogant. And get reading glasses and find out that im not whining about the board, im just stating a flaw that could possibly cause someone else to ruin ipods or any other usb device. I dont see how my posts havent really helped, but ok your brain works on a higher level than mine must be mr hawking.


Ok so yes im not whining about the board, and the fact that I can get more Ipods doesnt really say anything about money, more so about a good return policy and within 30days of buying it. but yeah lets get back on topic and stop flaming. :slapass:


Also - XBX and XBX2 look identical with a few minor changes, and the usb and 1394 ports on my board are in fact blue, not black. So I just posted the quickest pic I could find as I was on lunch when I made the post.

---

ignoring all the flaming here...

I'm looking at my board... and my USB headers are black while the firewire is blue...

mine are different colors... weird...

Quote:

---

Originally Posted by shimmishim

ignoring all the flaming here...

I'm looking at my board... and my USB headers are black while the firewire is blue...

mine are different colors... weird...

---

same here .. my usb are black and the other is blue as well

Quote:

---

Originally Posted by SPL15

In my case nope.

2507 is the most stable for me. 2330 and 2333 were not stable for me.

2507 and 2496 seam to be comparable, but 2507 seams to be more stable after several days of being on running folding or prime. 2495 would give me a random reboot after being on a for about 12 to 15 hours 2507 has fixed that for me.

After all my issues and stuff, I'm fully 100% rock solid several days of orthos stable at 413Mhz FSB on a E6600 which runs about 3.713Ghz. 4 GB of Crucial Ram running 1:1 which is DDR826 @ 4,4,4,10 with 2.24 Volts. I've been running Dual Prime95 for a week straight and no more degradation.

BIOS settings I've found to matter are:
C1E - Disabled
Watchdog - Disabled ( see explanation below why this WONT WORK!!)
Debug - Disabled
PCI Bus Latency - 64 (seams to help with USB devices and HDTV card lockups)
PCI express burn in - 101.x (seams to have helped lockups during games over default, anything higher than 101 gives me lockups)
Vmch - 1.7v actual from Vmod (I think I just have a weak and wussy 925X)
Vfsb - 1.425v (Helped with stability although I'm still testing this)
VT - enable (computer seams more stable after enabling even though it should have NOTHING to do with stability... maybe its my wussy chipset)
Vmem - 2.24 (usual stuff here nothing new)
Vcore - 1.45 (1.400 actual measured usual stuff here)

For the stupid "watchdog" setting on the XBX2, it will not work correctly on anything needing slightly hot voltage for the CPU WHY?? Well the XBX2, like most motherboards, has an initial powerup Vcore voltage that is per Intel specification of 1.1xx volts. Well if your CPU needs 1.45 volts to be stable at a certain overclock, then 1.1xx volts aint gonna get it stable enough to post correctly so the CPU clock will not be clocking and the watchdog will not get it's clock signal so the watchdog timer (thats on any commercially available computer circuit made by mankind) will pulse the reset line, by now the motherboard has given the CPU the higher VID voltage or cmos programmed Vcore voltage and now posts.

The intel "watchdog" is basically monitoring whether the watchdog is firing during post, if it fires, then it assumes that you have a poor overclock and gives you that stupid error. Watchdog timers are on any and every motherboard ever made by mankind and any and every embedded CPU, microcontroller, PIC, etc. It is nothing more than a system that watches for a clock signal and if it stops for a time length, it fires a reset signal in hopes of restarting the clock signal.

Using the watchdog "feature" on the XBX2 will limit your overclocks to a low enough setting that still enables the CPU to initially clock correctly at 1.1xx volts thus avoiding the watchdog from firing.

On other motherboards (p5b comes to mind) some people have problems with their system not posting until they hit reset. In this case, you are actually acting as a "watchdog timer". You physically see that there is no post which means there is no clock so you pulse the reset line and then things work because by the time you push the reset line, the Vcore is up where it needs to be in order to post.

Basically on the XBX2, the watchdog is a feature when it is disabled in the BIOS because the disabled setting allows the watchdog to reset the computer and give it another shot at posting when Vcore is higher, which is how a watchdog circuit should work in the first place. This concept unfortunately is not something new and has been around since before I was a fetus. Intel's watchdog setting basically tells you that the watchdog timer has fired a reset signal during post which means that the CPU isnt stable at the specified 1.1xx volts intial startup voltage. I basically see it as a way for Intel to limit Overclocking for a huge majority of casual overclockers / enthusiasts, thus making them want to upgrade to a better CPU in the future.

---

thanks .. this helped me alot ..
im still on the 2507 .. rev 505 here
so far im orthos stable at 3.6 on my e6400 where on the 680i boad i could only hit 3.5 .. so far 450fsb stable on the xbx2 with 1.46v actual .. 1.7 vmch .. 1.4 fsb .. 2.2 mem

Quote:

---

Originally Posted by SPL15

3.75 Ghz E6600 SL958 @1.4volts WC by Apogee
Intel D975XBX2 Rev 503 BIOS 2431 1066 Strap (This Mobo SUCKS!)

---

can i ask why you think this mobo sucks? im in the market...
you are the first person that i have heard doesnt like this board..
3.75 is pretty impressive!