the INtel P4 3.06ghz 800mhz FSB is in stock. here
:banana: :banana: :banana: :banana:
Printable View
the INtel P4 3.06ghz 800mhz FSB is in stock. here
:banana: :banana: :banana: :banana:
:toast: :toast:
15 multiplier, hmmmm......
Will need 300mhz fsb to hit 4.5
Might do it.
C
When the prices come down next week I'm thinking of moving back to Intel as I want a Canterwood. Ocworkbench posted Sandra scores for it :-
http://www.ocworkbench.com/2003/arch.../875sisoft.jpg
By the way shouldn't the 800Mhz FSB P4 be 3Ghz ?
The Intel forum here looks a bit thin, might change when Canterwood arrives.
Regards
Andy
if i have money, i´ll buy it.
But e prefer de xeon 3,0:D :D
Ahhh.......Quote:
Originally posted by zakelwe
When the prices come down next week I'm thinking of moving back to Intel as I want a Canterwood. Ocworkbench posted Sandra scores for it :-
http://www.ocworkbench.com/2003/arch.../875sisoft.jpg
By the way shouldn't the 800Mhz FSB P4 be 3Ghz ?
The Intel forum here looks a bit thin, might change when Canterwood arrives.
Regards
Andy
That's thoroughly ridiculous!
I want one.
C
hmmm this is interesting. Can't wait to test one out.
Chong
Question. Why does nobody else have one but this one store has it. Are they just not saying that it is for pre-order?
Chong
A store in Germany seems to have it in stock since friday. Only 517€ :D
http://www.overclockers.de/produkt.php?pid=1027&kat=127
Bah does the red dot mean out of stock ? i cant read German
Yep, out of stock.Quote:
Originally posted by toofast
Bah does the red dot mean out of stock ? i cant read German
This thing is SOOOOOOOOOOOO Sweet!
New P4 dream rig is based on IC7-G, 2 - WD 10k SATA Raptor's, a shiny new 3.2/800, 2 x 256mb Geil PC3700 Platinum and a Radeon 9800 Pro!
C
The ASUS P4C800 aint bad either..
Doing 3DMark2001SE at 294FSB :D
being a beta tester has its pluses.. :D
im waiting for the abit board myself.
$500?...now I remember why I play with AMD...:rolleyes:
I would mention that the equivalent amd part costs the same, but it appears the Barton 3000+ is only $340 now ... but yesterday they were the same!!Quote:
Originally posted by BrainStorm
$500?...now I remember why I play with AMD...:rolleyes:
Nice FSB speed !Quote:
Can you run the FSB and memory speeds totally independantly, or are there ratio's ie 3:2 etc ? Is there a performance hit not running sync FSB and memory
Regards
Andy
Dunno its not my score or mobo :)
just found the result from the ORB
but I'd guess that its at 294FSB/235MHz mem (4:5 divider)
in any case the performance is really at 'the next level'. that 3529MHz P4 rig at 294MHz FSB is producing lobby test scores equal to a 4.1GHz P4 on RDRAM mobo 175 4X or 4.15GHz on i845PE at 180FSB/240MHz mem.
That is amazing! I'll be getting the Abit when it's available, but I only have a 1.6A to pair with it. The 1.6 does 3dmark 2k1 @ 3.1GHz with the stock cooler though, so I have high hopes for water/vapo cooling:D.Quote:
VR-Zone has a Canterwood review up and they maxed out the Chaintech board's FSB at 255MHz w/ an ES 3.0.
hmm
all the canterwood reviews leave me with some questions...
there aren't any dividers on the ram
However you can run fixed ram speeds like 266-333-400... or am I wrong???
Meaning if you have a 3.0-200 you run it at 200dcddr
however if you decide to overclock it, your ram will be the problem at 1:1 (as it won't frequently be doing 260 :) )
I've read you can run the 533 cpu's at dcddr 400 too
but doesn't that mean your ram will bump up to 210 when you up the fsb to 150? of is it fixed?
on vrzone they used inverse 4:5 dividers slowing the ram, but are there also dividers that can make a 3.06 @ 23*170 do 225 on the ram?
cuz if I have my new prommy and take my 3.06 all the way to 4 ghz, I would like to have the possibility to run that ram at @ least what it's qualified for, being 217 :)
does anybody know?
I mean, not everybody that has a 3.06 will buy a new cpu...
if you have a 3.06 that clocks all the way to 4 ghz, what can you do with it on a cangterwood board?
we definitely need better memory to take advantage of these outrageous FSB speeds. I mean These boards could handle DCDDR700 (~265FSB with a 3:4 ratio--This is achievable with sub-zero cooling and a 14x or 15x cpu) without issue, unfortunately I think current DDR-I memory has reached somewhat of a peak at DDR500. Mind you, Intel can still benefit from higher asynch memory speeds because their bus is still quad pumped.
the question is... can we async the bus on 133 mhz cpu's?
My 3.0C is in tomorrow. Only board I have to test with is the BH7. Actaully the CPU will be here in about 8 hours.
Chong
Heres my next Intel setup :)
Get me a nice 4gig P4 on this baby and it's 'seeyea wouldn't want to beyea' to AMD :D
http://www.abit.com.tw/abitweb/webjs...DEL_NAME=IC7-G
OPP
The ASUS, AOPEN and Intel CAnterwoods are already out there, Abit's offering will probably lag behind (as usual)
But I like it a lot, check my Avatar!
C
ABIT's is out in a week. :D
Chong
Very good news! :DQuote:
Originally posted by Chong345
ABIT's is out in a week. :D
Chong
I'll be using one with my 2.66 to see if the Canterwood architecture will help me hit 3.9+. 845PE has me up to 3.8+
With Prommie...maybe 4gHz...
C
Hehe. Good luck. I hope you get there. There might be limited quantities of the ABIT board though at first. I would rush to get one if you find one.
Chong
I just know I'm going to be the last guy here to get the IC7-G .. Damn canada sucks for getting the NEWEST I still can't get a 9800 pro until the 25th...Oh well, sit and wait I shall...
somebody care to answer my question ;)
can we async 133 mhz cpu's on the canterwood boards,
because running a 3.06 @ 166 fsb would happily need a divider to get some bandwhith...
I have no idea what the BIOS options are like for this board. From what I have heard there are divider options like the ones you are describing. As for which board, the person I was talking to said he could not comment on that.
Chong
I read the reviews and such but it's too much cabbage for me atm to go with canterwood :(
I'll probably wait for Springdale
I'm almost sure that there are dividers. I looked at screenshot of Abit's Bios on Geil's website and they showed an option to set dividers, just not which dividers. I would imagine that you would get async dividers from the Taiwanese chip makers to run rma faster than FSB. Surely, they have dividers to run it slower! Even intels board will have that.
As for Springdale: Isnt that geared less at the performance/enthusiast market and more for the mainstream and small business?
There are dividers to run the memory less than 1:1 (such as 5:4), but not to run more than 1:1 (like 4:5 or 3:4), as there really is no need to give the P4 more bandwidth than it can use (at 1:1, the memory and FSB bandwidth are equal).
a p4 can use more bandwhith at quad pumped 133 mhz then 133 mhz on the ram...
Single-channel, sure (FSB 133x4x8 is 4.2GB/s bandwidth, DDR single-channel 133x2x8 is 2.1GB/s). But dual-channel DDR at 1:1 matches the P4's FSB bandwidth exactly (FSB 133x4x8 is 4.2GB/s bandwidth, DDR dual-channel 133x2x8, x2, is 4.2GB/s). For the new parts, it's FSB 200x4x8 for 6.4GB/sec of bandwidth, tied together with dual-DDR 200x2x8, x2, 6.4GB/sec.
This is why you see Sandra typically use in the high-90%'s with single-channel DDR, because the memory is not providing the FSB with enough bandwidth (even at 4:5 or 3:4 single DDR ratios). Many people refer to this as 'efficient', when in fact, it is really showing that the memory bandwidth is quite insufficient to allow the FSB any breathing room. With dual-channel 1:1 (or RDRAM 4x), the mem bandwidth is more than sufficient, and you'll see the FSB only using in the 70%'s of the available bandwidth (the FSB can't use 100% of the bandwidth because of overhead from other operations).
Adding even more bandwidth at that point would do next to nothing (less than 1% difference, if not in fact SLOWER due to asycronous operation, as you see in nforce2's).
So all u guys are saying Abit IC7-G, Hoo-Ra, Abit IC7-G, Hoo- Ra, but what if the gigabyte or Asus Canters outperforms Abit are u still goin to remain Loyal to Abit IC7-G.
hehehe
well, we'll see, the asus looks extremely performant and hi-fsb minded
i'm just looking for my Epox! ;)
uhhmmmmQuote:
Originally posted by calantak
the question is... can we async the bus on 133 mhz cpu's?
no ? :D
the bus will always be multiplied by your multiplier, no way around that.
it seems Epox 4PCA3+ will be released to stores worldwide next thurday!
I can hardly wait! :)
any idea how much the 4PCA3 will cost? im guessing this is the one without raid and some other goodies.
Around 199€
The cheaper version would be about 30€ less.
heheheQuote:
Originally posted by JCviggen
uhhmmmm
no ? :D
the bus will always be multiplied by your multiplier, no way around that.
You got my point :)
I have an asus right here next to me, I'll try it out and see what I can do with it :)
Same here! :DQuote:
Originally posted by Raptor
it seems Epox 4PCA3+ will be released to stores worldwide next thurday!
I can hardly wait! :)
Thanks for the update Raptor, best news I've heard all day. :)
I like this one because I plan on running RAID5 with four drives. If I go the Asus or Abit route, I'll be into it almost double the price.
True the EPoX HPT374 PATA RAID5 is software controlled, but with a fast processor, the overall system performance should really rock.
You did notice that they added the P4 power connector back that they did without on the 4T4A. The Techs said it wasn't necessary, but popular demand proved otherwise.
I'd laugh if it takes the OC lead over Asus and Abit--not holding my breath, but it would be funny.
Okay, where on Gods green earth can we actually GET one of these chips? I got the money, just can't find the chip!!!!
Dog, you actually won't want to do that...you'll get better performance from two SATA drives and ICH5R than you would with a PCI card.Quote:
Originally posted by Dog_One
Same here! :D
Thanks for the update Raptor, best news I've heard all day. :)
I like this one because I plan on running RAID5 with four drives. If I go the Asus or Abit route, I'll be into it almost double the price.
True the EPoX HPT374 PATA RAID5 is software controlled, but with a fast processor, the overall system performance should really rock.
You did notice that they added the P4 power connector back that they did without on the 4T4A. The Techs said it wasn't necessary, but popular demand proved otherwise.
I'd laugh if it takes the OC lead over Asus and Abit--not holding my breath, but it would be funny.
Refer to my earlier post...
http://www.xtremesystems.org/forums/...293#post130293
Wow! Thanks Zroc. I had my mind made up and would have wasted some bucks--had no idea that the PCI bus posed that much restriction on throughput. That certainly explains why most 4 channel RAID cards use either 64-bit or the run at 66MHz.Quote:
Originally posted by Zroc
Dog, you actually won't want to do that...you'll get better performance from two SATA drives and ICH5R than you would with a PCI card.
Refer to my earlier post...
http://www.xtremesystems.org/forums/...293#post130293
So do you suppose the EPoX solution for 4 channel onboard PATA RAID is DOA?
Purty much...it has to use the PCI bus, as ICH5R is only two-channel.
It would still be useful for RAID 0+1 or RAID 5, if it's capable. But if you're like me, I'd rather use the three drives in a RAID 0 + one drive for Ghost/imaging backups scenario. Then you have the speed AND the backup.
Not sure what they were thinking with the design of this board. One site indicates GbE via the MCH (which should be CSA) only the board has a Broadcom GbE controller. The other site makes no mention of CSA or MCH Ethernet. Hmmm...
I was kind of hoping that the EPoX engineers found a way to connect the Highpoint controller to the CSA bus and then had to use the PCI GbE solution. Now I'm not sure what to think.
Maybe it will be a great overclocker. It seems to be the last one out of the shoot, it better have something going for it that the others don't.