• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

The Official Granite Bay Review Thread

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.
I don't think my CPU is completely at it's max. I suspect it's the mobo. Also, there is not any RDRAM that I know of that will hold up at 4X in the 160+ FSB range.

The Asus GB mobo looks pretty decent, but it don't have the serial ATA, probably the only thing I don't like about it. There are rumors that Abit might not produce the GB7, but I sure hope they do.
 
I know the Gigabyte 850e board goes up to 156fsb. But I do not think it locks the AGP/PCI bus, at least there is no option in the bios for it. I have my 1.8 running at 150fsb in one right now. It is a rock stable board though. I have gotten it up to 156fsb using easy tune 4 with it once I am booted into XP. Anything over 150fsb though, and my sound card goes crazy on me.

Buzzdog
 
i just read cloudkat's post about using ddr400 ram in these dual channel boards...i read on this site that the memory bandwidth matches the cpu bandwidth when on these new dual channel boards when they run synchronously. doesn't that mean that if the ram frequency is faster than the cpu fsb, the extra memory bandwidth is "wasted"?
 
shanus said:
i just read cloudkat's post about using ddr400 ram in these dual channel boards...i read on this site that the memory bandwidth matches the cpu bandwidth when on these new dual channel boards when they run synchronously. doesn't that mean that if the ram frequency is faster than the cpu fsb, the extra memory bandwidth is "wasted"?
Click to expand...

That's what they say but since the real bandwidth is never 4200mb/s anyway I'm not 100% sure but to use that extra would require some really complex chip but that's probably not even possible becasue of some other reason that it newer gets higher or so. I lost my thought here I'm sorry :p
 
Ok, I looked at a couple more reviews on the Asus P4G8X and I'm beginning to like what I'm seeing. NOTE: I was wrong in my last post about the Asus mobo not having serial ATA, it does. Other features I like are 8X AGP with adjustable AGP voltage, adjustable vcore up to 1.85v, adjustable DDR voltage and lots of RAM timing settings, and adjustable PCI/AGP bus frequency. If Abit drops the ball on their GB7, looks like I'll probably be getting the Asus.
 
batboy said:
adjustable vcore up to 1.85v
Click to expand...
The AnandTech review says it can be set to 1.975v and the TBreak review says 1.950v.

I wouldn't hurt my little Northy with that much juice though.:)
 
I would;)

Looks like the ASUS board is going to be pretty good, but I'll have to wait...I'm broke and I have other things on my X-mas list.
 
Lets say in the second quarter of next year intel has a GB 3.2 on a 667 FSB and AMD has a 2.5 on the 64 bit. How do you think they will compare , standard and OC.
 
Mikey7,

Lets say in the second quarter of next year intel has a GB 3.2 on a 667 FSB and AMD has a 2.5 on the 64 bit. How do you think they will compare , standard and OC.
Click to expand...

/shakes magic 8 ball...........the future looks uncertain at this time.


Serious though, it is hard to speculate on technology that is not even released yet. The best that you can do is decide whether you want to upgrade now, or wait because your system you have now is doing the job.



On a side note......

I went to the forums at OCworkbench. They had just completed a review of the Gigabyte GA-8INXP. In the review it stated that they were only able to overclock a 1.8a to 153 fsb. They had a discussion on their forums about the motherboard. I asked the reviewer Bluetooth what had been the highest overclock he had been able to get with the 1.8a on other motherboards. He tried to dodge the question the first time I asked it, gave an answer of the granitebay chipset is intended for a server set up and that they do not overclock. Kind of ironic coming from a sight that has overclock in its name????:rolleyes: So anyway, I ask the question again, He finnaly gives me a straight answer, the highest he had been able to oveclock that cpu in other motherboards was...........you guessed it, 153 fsb. So in his test on the review, the CPU was what was holding it back, not the chipset. I hope this clears up some misconceptions about whether or not the granite bay chipset will work for overclocking.

Buzzdog
 
Buzzdog said:
So anyway, I ask the question again, He finnaly gives me a straight answer, the highest he had been able to oveclock that cpu in other motherboards was...........you guessed it, 153 fsb.
Click to expand...
Hehe, he started getting a little defensive about using a CPU with such a (relatively) low maximum FSB speed after you grilled him Buzz.....
from the keyboard of Bluetooth
listen .. workstations boards are never meant for overclocking. this board might retail for USD 200 and above.
Click to expand...
I don't think Gigabyte intended the GA-8INXP to be a workstation board anyway - why the OC options and blue LED lights etc.? Gigabyte even call the GA-8INXP the 'Ultimate Intel Pentium 4-Processor Based Desktop Platform'.

Just because Intel intended the E7205 to be a workstation chipset doesn't mean board manufacturers can't make desktop boards using it.:)
 
Hehe, he started getting a little defensive about using a CPU with such a (relatively) low maximum FSB speed after you grilled him Buzz.....
Click to expand...

:D Well, I did not see what the big deal was. Also, I am skeptical about his review in general. If you look at the test system specs, he said he was using a 2.53. Then he says he overclcoked a 1.8. Seems fishy to me. Could we have another [H] coverup??? I doubt it. All I was trying to get to was, was the GA-8INXP holding the cpu back. I really wish that more reviews would state what the highest fsb the test cpu has ever hit. Even if it is just with air cooling.

An example would be. To test for the over clocking ability of a motherboard. They should ideally use the same CPU, Memory, HSF, and vcore every time. Then they should let you know what is the highest that the CPU has ever hit stable. Then you could make an honest assesment of how well a motherboard or new chipset does for over clocking. All of the benchmarks in the world dont mean squat unless you eliminate all of the variables down to just the component being tested.

Maybe Ed might take a read at this and impliment this as a testing standard for future reiviews. It just kind of ticks me off, that all of these other sights that are really flashy and covered with ad's get shots at the early reviews. When you have a sight like overclockers.com that seems to get the sloppy seconds so to speak. I have read that Ed has a strict policy on what he recieves for reviews and testing. I think it was all over the Gheil memory debate. People were questioning if other sights reviews had been cherry picked sticks of ram to make the reviews a ringer.

Oh well, at least here, I feel that between the sight admins and the comunity in the forums, we get to the truth.

Buzzdog
 
Ideally, reviewers should use unlocked chips to really find the limit of the boards they review. Abbas at TBreak has an unlocked P4 that runs up to about 235MHz FSB (940FSB) - perfect for taking a board to it's FSB limit. It's a shame not every reviewer has an ES chip tho.:(
 
VCORE adjustment P4G8X?

Just curious....

Have anyone actually altered the VCORE on the P4G8X?

Anand at anandtech seemed to indicate it changed in the bios but not in reality with a hardware monitor...

Has anyone actually does so?

Does anyone know if the newer CPU's will work with the Granite Bay boards when they come out?
 
Re: VCORE adjustment P4G8X?

wheelerd69 said:
Does anyone know if the newer CPU's will work with the Granite Bay boards when they come out?
Click to expand...

I'd say it's a fair bet, just so long as they remain skt. 478.
 
 
The fact that none of them have exceeded 190mhz fsb is a little scare.
If that is a maximum for the chipset so that no GB-board can get over 190fsb we can all forget about support for the upcoming 0.09 micron processors if they are to use 200fsb.
If I remember right, both Asus and Gigabit boards stopped at 190fsb and at least one of those testers were using an ES cpu.
I'm still gonna buy this board but this is a little disappointing.
 
Mikey7 said:
Lets say in the second quarter of next year intel has a GB 3.2 on a 667 FSB and AMD has a 2.5 on the 64 bit. How do you think they will compare , standard and OC.
Click to expand...

Its going to be very hard to get a straight answer to that one. The Opteron looks very interesting, but we always have to remember that it will only show of the full power when running applications and OS that are designed for 64 bit operation. In a 32 bit desktop environment, it is going to be like running a SMP rig on an application that does not support dual processors.

What I fear will happen is that someone will run a benchmark using a 64 bit OS and application, and compare that to another proc using the 32 bit version of the OS and app. This will cause much, much screaming among the fanboys but in reality it is comparing an apple to an aardvark.


BHD
 
 
You must log in or register to reply here.

Similar threads

don256us
(Un)Official Race Week Thread.
2
Replies
26
Views
966
dfonda
dfonda
Y
ASUS TUF RTX 4090 - My review
Replies
0
Views
188
yoadknux
Y
Niku-Sama
AM5 board suggestions?
Replies
16
Views
2K
EarthDog
EarthDog
R
Still Considering New Laptop
Replies
8
Views
917
DeadSmiley
DeadSmiley
Al_bundy
Alucryl - 3DFX - CNC - Retro pc
Replies
2
Views
467
EarthDog
EarthDog
Back