• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

MB compatible with Conroe?

Donnie: Price is so subjective that nothing I say would convince you that AMD is cheaper, so whatever.

That aside, AMD uses an LGA for Socket F because the twisted pins are necessary for higher speed signals, i.e. high speed buses. I don't recall AMD ever "bagging on" the idea of LGA, but if you could point to me a press release where they did, I'd be happy to read it.
 
robberbaron said:
Donnie: Price is so subjective that nothing I say would convince you that AMD is cheaper, so whatever.

That aside, AMD uses an LGA for Socket F because the twisted pins are necessary for higher speed signals, i.e. high speed buses. I don't recall AMD ever "bagging on" the idea of LGA, but if you could point to me a press release where they did, I'd be happy to read it.

Robberbarron: If the price is so subjective, why do you guys keep spreading this boarderline misinformation and opinion as Fact?

AMD did say LGA was not important and made their normal knock on Intel. Their knee jerk reaction to anything Intel has when AMD doesn't is like clockwork. Only later do they try to keep quiet and hope links aren't found or its forgotten about when they follow up with the same tech. Not only AMD but AMD friendly Webmasters and Fans a like back them in such stupidness. They did the same thing with SSE, Hyperhthreading, Netburst and hell, even Banias LOL! The sad part is Intel Pulls the same Crap. Read Anandtech (Official AMD Mouthpiece) on the first 5xx with Sc-775 to see what I mean? Same can be seen at Sudhian, Vans Hardware and other AMD fansites such as the AMD Zone where I think I first saw the negative comments.
 
Donnie27 said:
Robberbarron: If the price is so subjective, why do you guys keep spreading this boarderline misinformation and opinion as Fact?

AMD did say LGA was not important and made their normal knock on Intel. Their knee jerk reaction to anything Intel has when AMD doesn't is like clockwork. Only later do they try to keep quiet and hope links aren't found or its forgotten about when they follow up with the same tech. Not only AMD but AMD friendly Webmasters and Fans a like back them in such stupidness. They did the same thing with SSE, Hyperhthreading, Netburst and hell, even Banias LOL! The sad part is Intel Pulls the same Crap. Read Anandtech (Official AMD Mouthpiece) on the first 5xx with Sc-775 to see what I mean? Same can be seen at Sudhian, Vans Hardware and other AMD fansites such as the AMD Zone where I think I first saw the negative comments.
"You guys"? What group are you lumping me into now?
Again, show me where AMD made a press release saying LGA was not important, not hardware review sites that you believe are AMD subsidaries.

As for Intel pulling the same crap, I don't recall Intel ever saying anything negative about AMD. They merely just pretend AMD doesn't exist.
 
robberbaron said:
"You guys"? What group are you lumping me into now?
Again, show me where AMD made a press release saying LGA was not important, not hardware review sites that you believe are AMD subsidaries.

As for Intel pulling the same crap, I don't recall Intel ever saying anything negative about AMD. They merely just pretend AMD doesn't exist.

The you guys I'm talking about the are the whole "AMD is cheaper" croud. Lump, well yeah! Please show me one where they didn't, you provide a link or links?
 
The burden for proof of that is on you, sorry. Learn to debate before next time, ok?
 
robberbaron said:
The burden for proof of that is on you, sorry. Learn to debate before next time, ok?

The burden to disprove something I stated is on you LOL! Like the other guy said, stick to posting about Conroe.
 
$BangforThe$ said:
The 975 is infact suppose to work with the conroe Cpu but a bios update will probably be needed. There seems to be a problem with the power regulators between the P4's and conroe . So there's a good chance that the 965 will be required. Dam intel . But I do see why the power requirements would be differant but that being said . P4 should require much more power than Conroe so that A bios update may be all that is required for the 975 chipset to work. :confused:

http://www.theinquirer.net/?article=29599

The firm is also pressing on with re-branding of its microprocessor estate, with both Merom notebook chips and Conroe desktop chips becoming branded as Intel Core processors. As we revealed earlier this week, Conroe desktop chips will be numbered as E6000 and E4000, the latter category being for 2MB cache, 800MHz bus Conroes and the former for 1066GHz, 4MB and all that.

Most stories/reports I've read haven't mentioned i975 at all. In this case, no news is bad news. OTOH, it seems that i965 and i985 might be well worth the waite as well. They will support newer features like;support VT and AMT, Matrix Storage Tech, and use HDMI.
and so on. We'll also see 3 FSB's 800, 1066 and 1333MHz.
 
I think it's ridiculous personally... I bought a "975x" board, and I will need a "965" to run it? That's a model down! :mad:

As I said earlier in the thread... Intel needs to STOP with this 1:1 crap. What was the point of releasing the 975x in the first place? The 955 and 945 can run the Preslers and Cedar Mills just fine. If it can't run their next gen platform, it was a useless release.
 
The 955 and 945 can run the Preslers and Cedar Mills just fine. If it can't run their next gen platform, it was a useless release.

Intel is also engaged in the chipset business :D
 
steviep said:
I think it's ridiculous personally... I bought a "975x" board, and I will need a "965" to run it? That's a model down! :mad:

As I said earlier in the thread... Intel needs to STOP with this 1:1 crap. What was the point of releasing the 975x in the first place? The 955 and 945 can run the Preslers and Cedar Mills just fine. If it can't run their next gen platform, it was a useless release.

I think and CAN'T prove it but Intel shipped i975 early and without all of its planned features. Now i975X becomes i985 instead of getting an add-on XEX or etc.. I want Conroe but I'm not buying anything from Intel until this is settled. Might easily get i985X though.
 
I know you work there empoy, and I know they need to make money off chipsets. But c'mon, you can admit that it's getting a bit ridiculous. It's the same socket, I'm sure it can be made to work. This 1:1 release of CPUs and motherboards is NOT a necessity (see: P2/P3/AMD), especially when it's the same socket. I understand why the 915 won't work with Conroe, maybe even the 925! But everything beyond that should be made for compatibility. I just wanted to get a 975 motherboard, get a cheap 930 now to tide me over, and later this year upgrade to a Conroe. But I refuse to have to buy another motherboard for it. Especially one that's a model down :p

PS Donnie, I heard the same thing about the 975. But why release it in the first place, then? If there is nothing really new to offer (aside from Crossfire) over the 955? Or why not wait an extra month before you release it so that you can include all that stuff like ICH8? Get what I'm trying to convey? I'm pissed.
 
steviep said:
Or why not wait an extra month before you release it so that you can include all that stuff like ICH8? Get what I'm trying to convey? I'm pissed.

This comment right here tells me that you don't know much about how this stuff works. ICH8 is still in validation stage and probably won't come out till Q3 at the earliest. That's a bit more than a month. You'll see broadwater (next gen MCH) boards without ICH8 before that most likely..... and you can run Conroe on those, but it's a "value" board. Now... 975 really isn't anything different than 925XE, it's a performance version of the existing platform. It's pretty easy to turn these out with little validation since most of it has already been done. Does anybody remember the difference between 925 and 925XE? Basically just an increased FSB to sell ONE processor. That's it. Tell me how this is any different?

What I don't get is why everybody is getting all pissy? Is it because they gave 975 a new number? Would you have been happier if they called it 955XE? Seriously.... plan your upgrades. If you know that you want a conroe and know that it's coming out Q3.... then wait. Or if you can't wait build a system now with a 9xx and be happy that you have a pimp shit system. This is the price you pay if you want the latest and greatest tech. These yearly board changes don't affect most people since most people are happy upgrading every 2-3 years.
 
steviep said:
I know you work there empoy, and I know they need to make money off chipsets. But c'mon, you can admit that it's getting a bit ridiculous. It's the same socket, I'm sure it can be made to work. This 1:1 release of CPUs and motherboards is NOT a necessity (see: P2/P3/AMD), especially when it's the same socket. I understand why the 915 won't work with Conroe, maybe even the 925! But everything beyond that should be made for compatibility. I just wanted to get a 975 motherboard, get a cheap 930 now to tide me over, and later this year upgrade to a Conroe. But I refuse to have to buy another motherboard for it. Especially one that's a model down :p

PS Donnie, I heard the same thing about the 975. But why release it in the first place, then? If there is nothing really new to offer (aside from Crossfire) over the 955? Or why not wait an extra month before you release it so that you can include all that stuff like ICH8? Get what I'm trying to convey? I'm pissed.

I couldn't agree more! Like the other guy joked about, Intel sells chipsets as well.;) I'd rather see a delay for compatability reasons than un-needed chipsets. Especially when those chipsets aren't compat with with a wider range of products. Stable Platform my #%$^@&*$^$ @$$!
 
Donnie27 said:
I'd rather see a delay for compatability reasons than un-needed chipsets. Especially when those chipsets aren't compat with with a wider range of products. Stable Platform my #%$^@&*$^$ @$$!

LOL.... then you call up Craig Barret and tell him you'll give him 400million advance on an order that you plan to make if Intel will release these chipsets on your time table, trust me it'll work since that is basically what Dell does. Dell, HP, IBM, etc. are the ones who are driving these chipset dates..... NOT Intel. They are the ones saying.... "Hey, we want these features" And we say.... "pay up bitch and we'll give them to you" And they're like "word."


Ok... maybe it's a bit more complicated than that, but you get the idea. Sorry... but the pimply faced fan boy on [H]ard Forums isn't driving the market. Dell, and by extension their customers, are driving the market. Does that make sence?
 
Poncho said:
This comment right here tells me that you don't know much about how this stuff works. ICH8 is still in validation stage and probably won't come out till Q3 at the earliest. That's a bit more than a month. You'll see broadwater (next gen MCH) boards without ICH8 before that most likely..... and you can run Conroe on those, but it's a "value" board. Now... 975 really isn't anything different than 925XE, it's a performance version of the existing platform. It's pretty easy to turn these out with little validation since most of it has already been done. Does anybody remember the difference between 925 and 925XE? Basically just an increased FSB to sell ONE processor. That's it. Tell me how this is any different?

That's exactly what I'm complaining about! They made a completely new motherboard to sell one processor... WTF? My 975 board is a tuned 925xe? Well fantastic, why didn't they just KEEP THE 925?!!! lol. Make the CPUs backward compatible a bit more maybe?

What I don't get is why everybody is getting all pissy? Is it because they gave 975 a new number? Would you have been happier if they called it 955XE? Seriously.... plan your upgrades. If you know that you want a conroe and know that it's coming out Q3.... then wait. Or if you can't wait build a system now with a 9xx and be happy that you have a pimp shit system. This is the price you pay if you want the latest and greatest tech. These yearly board changes don't affect most people since most people are happy upgrading every 2-3 years.

I'm not knocking on you personally, since I know you're a good guy and share a lot of useful information with us. But as you can tell, I am quite angry that the Conroe CAN work in the 975 (as you've said) but it likely will not. Yes I would've been happier if they named it properly (955xe) because then at least the general consumer knows he's not getting ripped off. ("why does the 965 run the new chip, but not the 975 I have?"). I needed a new machine because my current *P3* machine is slowly becoming unusable for the applications I need. I couldn't wait longer, despite the fact that I wanted to. So I purchased their latest platform (975) hoping that perhaps they would stop with this 1:1 crap and release a CPU that runs on the same damn socket that actually works on a motherboard more than 3 months old. I'm the kind of person that does my major upgrades every 2-3 years (with small incremental updates every year or so), as you can tell, and I am rightfully pissed that my upgrade path is extremely limited. It didn't used to be this way.
 
steviep said:
That's exactly what I'm complaining about! They made a completely new motherboard to sell one processor... WTF? My 975 board is a tuned 925xe? Well fantastic, why didn't they just KEEP THE 925?!!! lol. Make the CPUs backward compatible a bit more maybe?

You misunderstood what I said. Your 975 IS NOT a "tuned" 925. 975 is based off the 955 chipset, and is a performance sku of that. I was making the comparison to the different 915/925 versions.

steviep said:
Yes I would've been happier if they named it properly (955xe) because then at least the general consumer knows he's not getting ripped off. ("why does the 965 run the new chip, but not the 975 I have?").

I kinda agree with you here, this number scheme is a bit confusing. They should have called 975 chipset 965 and left 975 for Broadwater. But whatever.... not really all that difficult to remember. The only people who know the chipset numbers are people that i would hope, stretching I know, but I would hope they would understand what numbers go where.

steviep said:
I needed a new machine because my current *P3* machine is slowly becoming unusable for the applications I need. I couldn't wait longer, despite the fact that I wanted to. So I purchased their latest platform (975) hoping that perhaps they would stop with this 1:1 crap and release a CPU that runs on the same damn socket that actually works on a motherboard more than 3 months old. I'm the kind of person that does my major upgrades every 2-3 years (with small incremental updates every year or so), as you can tell, and I am rightfully pissed that my upgrade path is extremely limited. It didn't used to be this way.

Ok... so you went from a P3 to a presler based system and aren't happy? Dude.... you have a pimp system. Be happy with it. The system that you have will last you another couple years at least. I'm just sensing that you have a desire to throw your E-Penos on the table by having the latest and greatest hardware. Sorry... but this line of reasoning is expensive. I know... because I have done it. I'm the idiot who paid almost 800 dollars so I could have 2 36 gig raptors the week they came out. I'm the idiot who will spend 700 dollars on the latest graphics card, not because it's that much better than what I currently had, but just because its the best. That's just the price paid to have the biggest wang. Deal with it.

And you need to realize that there is more to processor/chipset compatibility than the socket. Guess what.... my presler based xeons will fit in a LGA775..... but it'll probably blow something up if I did it. But they should work since it's the same socket right?
 
Poncho said:
LOL.... then you call up Craig Barret and tell him you'll give him 400million advance on an order that you plan to make if Intel will release these chipsets on your time table, trust me it'll work since that is basically what Dell does. Dell, HP, IBM, etc. are the ones who are driving these chipset dates..... NOT Intel. They are the ones saying.... "Hey, we want these features" And we say.... "pay up bitch and we'll give them to you" And they're like "word."


Ok... maybe it's a bit more complicated than that, but you get the idea. Sorry... but the pimply faced fan boy on [H]ard Forums isn't driving the market. Dell, and by extension their customers, are driving the market. Does that make sence?

Yes, it makes very good sense. If I were talking about Dell, HP and the rest, I would have said something similar, get it? I know full and well how it works. It's about time for this market to get something like a Stable Platform promise and not BSed! Oh and this 49 year old (Feb 1, 1957) man has played with this crap since 1983 LOL! I doubt the OEMs asked for 3 different i925s=P :rolleyes: I only look for something to last me as long as my i875 did. Even i955 makes sense, Dual Core support. i975X doesn't, 6 month lifespand is half that of Sc-423. Maybe for PR reasons Intel should have named it i958XE LOL some more!
 
Donnie27 said:
Oh and this 49 year old (Feb 1, 1957) man has played with this crap since 1983 LOL! I doubt the OEMs asked for 3 different i925s=P :rolleyes:

Ok... the pimply faced comment was a joke. Should have put a smiley in there. :D

As for what the OEMs asked for.... you'd be surprised. You also have to take into account all the Board manufactures as well. They all have different wants and needs. If enough of them ask for it.... they'll probably get it. This doesn't only fall on Intel, and I think people fail to realize that. Intel provides a service, that's where most of the money is made for us. We ask our customers what they want and try our best to give it to them. You buy a board from Asus..... guess what, Intel didn't build it. If it doesn't work, it's probably Asus' fault. But since it's an "Intel" board then people blame Intel. Works that way for just about everything. I'm not saying that Intel doesn't drive anything, they drive a lot, but just that people need to realize there is a bigger picture here.
 
Poncho said:
Ok... so you went from a P3 to a presler based system and aren't happy? Dude.... you have a pimp system. Be happy with it. The system that you have will last you another couple years at least. I'm just sensing that you have a desire to throw your E-Penos on the table by having the latest and greatest hardware. Sorry... but this line of reasoning is expensive. I know... because I have done it. I'm the idiot who paid almost 800 dollars so I could have 2 36 gig raptors the week they came out. I'm the idiot who will spend 700 dollars on the latest graphics card, not because it's that much better than what I currently had, but just because its the best. That's just the price paid to have the biggest wang. Deal with it.

And you need to realize that there is more to processor/chipset compatibility than the socket. Guess what.... my presler based xeons will fit in a LGA775..... but it'll probably blow something up if I did it. But they should work since it's the same socket right?

Of course I realize that. But is there really any great reason why Conroe SHOULDN'T work with 975? As you've said yourself, it does work in the test phase you tried it in, but won't in the end. Why do they do this? To sell chipsets/motherboards, of course :rolleyes: There should, by all accounts, be a BIOS upgrade for these 975 boards that should make it work. Hell, they should make it work in the 94/55. I understand if the Extreme Edition doesn't work, since my board only goes up to 1066fsb... but the regular Conroes definately should.

I really wanted to wait for Conroe, but I couldn't since I needed a system now. Yes, the one I just built is fantastic, but I have heard so many good things about the Conroe architecture that I was planning on making that my yearly upgrade next year. The fact that I most likely will not be able to without a big investment (motherboard as well) kind of irks me a bit, that's all.
 
steviep said:
But is there really any great reason why Conroe SHOULDN'T work with 975?

Ummm..... Yea, it hasn't been validated yet. Nobody can come out and officially say that it works, when they haven't put it through the paces yet. And during this process there might be little tweeks here and there to the board design to allow Conroe to work. You can't EXPECT a board that you purchased during it's first production rev. to work with a processor that isn't due out for another 8 months. It would be nice if it did....but it's not guaranteed. Now... another thing is that you are also going to a completely different architecture. This isn't a drop in thing, there are a lot of changes that have to be made to get it to work. Look at the big changes for the past few platforms


845 - Added DDR support and 533 FSB/HT.
865/875 - Added 800 FSB, dual channel memory, Prescott support, ICH SATA
915/925 - Added 1066FSB (925xe), DDR2 - 533, PCIe
945/955/975 - Added Dual Core support, Dual Graphics, DDR2-667, SATA Gen.2

These are all pretty big changes from platform to platform. And in reality it makes the upgrade path a lot more manageable. If every 3 years you had to upgrade 95% of your hardware people would be bitching about that. But instead they do big changes, most with some type of backward compatibility, a little at a time.
 
Poncho said:
Ok... the pimply faced comment was a joke. Should have put a smiley in there. :D

As for what the OEMs asked for.... you'd be surprised. You also have to take into account all the Board manufactures as well. They all have different wants and needs. If enough of them ask for it.... they'll probably get it. This doesn't only fall on Intel, and I think people fail to realize that. Intel provides a service, that's where most of the money is made for us. We ask our customers what they want and try our best to give it to them. You buy a board from Asus..... guess what, Intel didn't build it. If it doesn't work, it's probably Asus' fault. But since it's an "Intel" board then people blame Intel. Works that way for just about everything. I'm not saying that Intel doesn't drive anything, they drive a lot, but just that people need to realize there is a bigger picture here.

No offense taken, hell, you've been pretty cool with me. For the record, I HEAVILY deffended Intel over the whole FMB Prescott/i875/65 schlock! Same Deal with PAT (some folks had problems when Asus and ABit enabled it on i865). I also defended Intel vs RAM makers rebadging PC2700 as PC3200, saw it first hand with my old i875. Some of the mobo makers complained about too many chipsets, on the Inquirer and Digitimes. Intel is my Favorite Chipset maker=P Again, my complaints have more to do with the last two i925s and i955 than any *one of them (Chip Sets) as a whole.

On another note, the free pass nVidia is given on holding SLI to itself is BS as well. Will Conroe compat nForce boards be the only ones to support nVidia's SLI? If Intel were doing this, they'd be hung and burned by rioting mobs!
 
Donnie27 said:
Will Conroe compat nForce boards be the only ones to support nVidia's SLI? If Intel were doing this, they'd be hung and burned by rioting mobs!

Yea, that one is kinda fucked up. They will be the only "official" boards that support both, but there will be hacked drivers that allow it. They've pretty much said "Fuck you" to intel.
 
Poncho said:
Yea, that one is kinda fucked up. They will be the only "official" boards that support both, but there will be hacked drivers that allow it. They've pretty much said "Fuck you" to intel.

Yup!

Let's look at the Conroe thing this way?

I don't think many are complaining about i845 SDRAM, to DDR or 848 and danged sure not i875 or i865. Many wish i915 through i975 was equally well thought out. Conroe stampped out when? SATA 2, ICH8 and etc..? What I meant was maybe for PR reasons, is that this i975 could have been given an i848 like title of i958 or even i960. i975 gives it a Conroe Generation title and make folks think it will be Conroe compat since that number is higher than the Conroe compat i965 as per what ANY Intel user expects.. ala i875/865, i925/i915, i955/945 and many would just expect i975/i965 in that same mold.

Even if it is compat, i975 will be less Advanced than i965. Surely someone noticed this? MAYBE i960 wouldn't have generated much pissing and bitching about it if Conroe is compat or not. Seen many threads like this about i955 being Conroe compat? Does that sound unreasonable to you? Transistion Mobos and Chipsets should be given transistional names.

Donnie27, very happy Intel user for at least the last 14 years.
 
If Conroe is working in 975 (preprod) boards now.
I'd be willing to be money, Asus or Abit will release a BIOS to support it, when the chip comes out.

We're still talking about months away. Look what Asus did to support PentiumM on s478 (CT-479 adapter). If there is anyway for them to make it work they will.

And considering this early in the stage, the samples are working. I'd bet they will work on release.

The only problem I see is the ConroeXE might not work in all 975 boards b/c of the 1333FSB, as it would require being overclocked. And not all boards will cut it.
 
Poncho said:
845 - Added DDR support and 533 FSB/HT.
865/875 - Added 800 FSB, dual channel memory, Prescott support, ICH SATA
915/925 - Added 1066FSB (925xe), DDR2 - 533, PCIe
945/955/975 - Added Dual Core support, Dual Graphics, DDR2-667, SATA Gen.2

These are all pretty big changes from platform to platform.

Those ARE all big changes. But you know what is sad!?

A 3.4EE on an 865/875 motherboard with a 6800GT
will perform as well as a
661 Cedarmill on a 975 motherboard with a 6800GT
(and similarly as a 955XE, unless the app is multi-threaded)

So a 2 year old chip & motherboard, with the same graphics card (though different interfaces AGP vs PCIe) will perform approx. the same as the latest & greatest CPU + chipset generally speaking.

I think the problem is Intel us holding back on the clocks of the new P4's & features of motherboards b/c they don't want Conroe + new motherboards to be slower than the P4 when its released. B/c from everything I'm seeing Cedar Mill could be easily sold stock at 4.2+Ghz right now.
 
chrisf6969 said:
If Conroe is working in 975 (preprod) boards now.
I'd be willing to be money, Asus or Abit will release a BIOS to support it, when the chip comes out.

We're still talking about months away. Look what Asus did to support PentiumM on s478 (CT-479 adapter). If there is anyway for them to make it work they will.

And considering this early in the stage, the samples are working. I'd bet they will work on release.

The only problem I see is the ConroeXE might not work in all 975 boards b/c of the 1333FSB, as it would require being overclocked. And not all boards will cut it.

I wouldn't take that bet with you ;)

If it's just a matter of a BIOS flash I'm sure they'll do everything they can to support Conroe. I personally would rather have the new board anyway. Yet, I do agree with those wanting to hold onto their excellent i975 boards out there now. Hell, it's no surprise why nVidia is holding back support.
 
chrisf6969 said:
Those ARE all big changes. But you know what is sad!?

A 3.4EE on an 865/875 motherboard with a 6800GT
will perform as well as a
661 Cedarmill on a 975 motherboard with a 6800GT
(and similarly as a 955XE, unless the app is multi-threaded)

So a 2 year old chip & motherboard, with the same graphics card (though different interfaces AGP vs PCIe) will perform approx. the same as the latest & greatest CPU + chipset generally speaking.

I think the problem is Intel us holding back on the clocks of the new P4's & features of motherboards b/c they don't want Conroe + new motherboards to be slower than the P4 when its released. B/c from everything I'm seeing Cedar Mill could be easily sold stock at 4.2+Ghz right now.

661 vs 3.4EE is a very bad example though. Because if the cooler running, less power using, and much cheaper CM 661 can hang, it is a Much better deal. Tom's hardware has a 3.4 Prescott vs 3.4GHz Cedar Mill comparo.

http://www.tomshardware.com/2005/10/07/a_sneak_peak_at_intel/
 
Cedar Mill @ 4.7 ghz FTW!

seriously... once i get the money im hooking up my triple radiator water system to one of these things and I'll be able to just skip the 1st generation of Conroe just fine :)
 
I've heard 945, 955, 975... new to come. So current chipsets support it, but i'm not sure how efficient they'll make it.. probably won't be full potential. About those who say that the ConroeXE's have 1333 FSB, Intel makes motherboards... they'll get on top of that before they sell a processor, lol, don't worry about that
 
945 and 955 support for Conroe? I'd doubt that, but we will see.
 
Donnie27 said:
661 vs 3.4EE is a very bad example though. Because if the cooler running, less power using, and much cheaper CM 661 can hang, it is a Much better deal. Tom's hardware has a 3.4 Prescott vs 3.4GHz Cedar Mill comparo.

http://www.tomshardware.com/2005/10/07/a_sneak_peak_at_intel/

I used a 3.4EE b/c its an "old" chip with 2Mb of (L3) cache, so its a fair comparison..... 2Mb vs 2Mb

The CedarMill process is half the size! 130nm / 2 = 65nm ! And they haven't really done ANYTHING to make it faster at the same speed. (Clock for clock)

AND they're only releasing chips < 400Mhz faster than these SUPER OLD chips. 3.46Ghz vs 3.8 Prescott

Just saying.... its a little disappointing, with ALL of those fancy new features, like PCIe, 2 die shrinks, SATA, faster FSB's available (but not being used), etc... and the general speed of a 2 year old system is the same as the latest & greatest..... (if you have equal videocards)

The CPU / Motherboard / Ram combination has made practically 0 improvements in the past 2 years. Besides "blue crystals".

They should be releasing at least 4++Ghz 6x1's right now.
 
chrisf6969 said:
I used a 3.4EE b/c its an "old" chip with 2Mb of (L3) cache, so its a fair comparison..... 2Mb vs 2Mb

The CedarMill process is half the size! 130nm / 2 = 65nm ! And they haven't really done ANYTHING to make it faster at the same speed. (Clock for clock)

AND they're only releasing chips < 400Mhz faster than these SUPER OLD chips. 3.46Ghz vs 3.8 Prescott

Just saying.... its a little disappointing, with ALL of those fancy new features, like PCIe, 2 die shrinks, SATA, faster FSB's available (but not being used), etc... and the general speed of a 2 year old system is the same as the latest & greatest..... (if you have equal videocards)

The CPU / Motherboard / Ram combination has made practically 0 improvements in the past 2 years. Besides "blue crystals".

They should be releasing at least 4++Ghz 6x1's right now.

You mean you don't overclock? Can any of the older mobo reas 1333 to 1400MHz? Stock speed becomes almost moot.

Still not the same. You're talking about L2 with 512K + L3 with 2048K vs. 2048K, simply put thats 2.5MB vs 2MB. My way used 2MB vs 2MB L2s and NO L3.

It is also easy to used software that DOESN'T show the newer systems advantages over the older system. Limit tests to older software or software that they both suck at running.
Some apps love quick take-offs, others love sustained bandwidth.

Even if performance us absolutely dead even, the new systems run cooler and uses less power, something that was badly needed. Some even see it as a fix for older systems.

You, I or NO one else knows how high Cedar Mill will go, so please don't give us numbers you can't back up. NO ONE thought Northwood would ship at 3.4GHz for example. Any overclocker could tell you Intel could have shipped 4GHz a long time ago.

Last be not least, Cedar Mill is a transistional Product mainly meant to validate 65nm. The true next gen Product is Merom, Conroe and Woodcrest. AMD is doing something similar by trying out its new tech on 90nm first before going to 65nm sometime next year.
 
many 875/865's could hit 1200FSB, but were limited past that by some built in PCI/AGP bus problem past that. My Abit IC-7 hit 1200FSB easily.

The 3.4EE might have total 2.5Mb, but since the L2 cache is repeated in the L3, its still only an effective 2Mb cache. And the L3 has a higher latency being further down the line. So if anything the 3.4EE has a disadvantage... but is the closest thing to the current 2Mb L2.

I know the Cedar Mills will blow Northwood & Prescott away when overclocked.

But the average user buying a Dell, is getting virtually the same performance now as he did 2 years ago. Thats all I'm saying is it kinda sucks that 2 years later, they're releasing shit that performs virtually the same as their 2 year old chips.

I think its b/c they're holding back CedarMill & Presler speeds to make sure when Conroe hits the scene it comes with a decent improvement. They don't want another P4 williamette launch with the P3's outperforming them.
 
chrisf6969 said:
many 875/865's could hit 1200FSB, but were limited past that by some built in PCI/AGP bus problem past that. My Abit IC-7 hit 1200FSB easily.

The 3.4EE might have total 2.5Mb, but since the L2 cache is repeated in the L3, its still only an effective 2Mb cache. And the L3 has a higher latency being further down the line. So if anything the 3.4EE has a disadvantage... but is the closest thing to the current 2Mb L2.

I know the Cedar Mills will blow Northwood & Prescott away when overclocked.

But the average user buying a Dell, is getting virtually the same performance now as he did 2 years ago. Thats all I'm saying is it kinda sucks that 2 years later, they're releasing shit that performs virtually the same as their 2 year old chips.

I think its b/c they're holding back CedarMill & Presler speeds to make sure when Conroe hits the scene it comes with a decent improvement. They don't want another P4 williamette launch with the P3's outperforming them.


I agree with the first part of the post totally... ppl buying computers for the most part aren't getting anything better than ppl did 2 years ago! Sheesh, tech has really slowed...

About that last part...hmmm, I dunno... do other ppl think an OCed Cedar Mill would be faster than the Conroe when its released (running at stock speeds..) Honestly, I doubt anyone can really answer that question, but I seriously doubt that we'll get another WIlliamette outta the Conroe.
 
many 875/865's could hit 1200FSB, but were limited past that by some built in PCI/AGP bus problem past that. My Abit IC-7 hit 1200FSB easily.[/QUOTE]

Dewd, 1200 is NOT 1333 or 1400MHz=P

chrisf6969 said:
The 3.4EE might have total 2.5Mb, but since the L2 cache is repeated in the L3, its still only an effective 2Mb cache. And the L3 has a higher latency being further down the line. So if anything the 3.4EE has a disadvantage... but is the closest thing to the current 2Mb L2.

Nope, some sights questioned if that 512MB was repeated in L3. Cedar Mill will not come close to the P4EE price points. Makes no sense to compare it to something that costs $$$$ more. See below?

chrisf6969 said:
I know the Cedar Mills will blow Northwood & Prescott away when overclocked.

That's what it is supposed to do, NOT compete with P4EE, that job is reserved of 955XE and etc..... Cedar Mill is meant for mainstream and nearer to low-end than the top. Yes, for Dell, HP and Gateway for Business Workstations for secretaries or etc.. Finally for oerclocker in this market looking for bang for the buck.

chrisf6969 said:
But the average user buying a Dell, is getting virtually the same performance now as he did 2 years ago. Thats all I'm saying is it kinda sucks that 2 years later, they're releasing shit that performs virtually the same as their 2 year old chips.

No they will NOT, that's mainly do to sucky built in Video that sucks less, most importly faster SATA-2 HDDs, improved I/O as whole and other things. Folks like us who might tweak the hell out of out our DIYer rigs will noticed less than some Newbie using a 2 or 3 year Dell system with UATA-100 with a 2MB cache. Their new one now using a SATA-2 and etc... I can easily tell the difference between my Seagate 160GB SATA-1 8MBc and my WD 250GB SATA-2 16MB cache. I too own an IC7-G Max and I tried this Drive on it before installing it on my nForce SLI A8N. I can easily tell the difference.

chrisf6969 said:
I think its b/c they're holding back CedarMill & Presler speeds to make sure when Conroe hits the scene it comes with a decent improvement. They don't want another P4 williamette launch with the P3's outperforming them.

I disagree 100% there's ZERO need to hold Cedar Mill or Presler back. Comparing a first stepping to Mature tech is BOGUS IMHO. Conroe is going to kick the crap out of Intel's current offerings. There were apps that the Willy killled the P3 on. Intel advertised P4 as "The Center of your Digital World". It did that perfectly. Too bad many Webmaster didn't bother to run the "Intel Launcher software 1.21". THIS WAS OFF THE SHELF software optimized for SSE2. You might have read too many Anantech reviews when he still had FOP code enabled in the BIOS causing a 17% performance hit. Or didn't bother to run streaming apps in general. Or Pretending he couldn't find USAF, Incoming Forces, DIVX with Intel's SSE2 IDT or even patched LAME, sheesh!
 
my point was that many 875's did 1200Mhz easily with the right CPU. The only problem for hitting more than 1200 was some built in limitation that made things go screwy for the AGP once you went over 300FSB.
Conroe is slated for 1066, with 1333 for the extreme edition.

the 512 is repeated in the 2Mb. P4's repeat the smaller cache in the larger. They're inclusive. AMD's chip's caches are exclusive so you can add them.

8mb harddrives have been out for ages! (you're going back too far) Granted some crappy OEM systems came with only 2Mb drives a few years back.

But I was talking about CPU / ram / motherboard, anyway That the problem, they haven't been making much progress the past 2-3 years. We are in the processors forum, not the storage/hd forum.

I'm an Intel f4nboy, but you've got to admit their stock processors have made VERY little improvement over that past 2 years.

I completely skipped the 90nm process for Intel, b/c there was nothing worthwhile. Finally with Cedarmill and Presler there's a reason to upgrade from an old Northwood/875 system b/c you can really overclock the sh1t out of these new 65nm processors. (but if you're buying a stock system, you're getting about the same performance as something 2 years old. )
 
my point was that many 875's did 1200Mhz easily with the right CPU. The only problem for hitting more than 1200 was some built in limitation that made things go screwy for the AGP once you went over 300FSB. Conroe is slated for 1066, with 1333 for the extreme edition.

No, it didn't do 300 MHz FSB that easily. Chipsets are just like CPUs and all of them didn't do 300. On the New ones we're taking about 266 Standard and 333MHz FSB highend. Intel will always build some kind of headroom into these bad boys so we're talking at least 1466MHz FSB. These might even reach 1600MHz effective or 400MHz base. Still wodering why AMD is going to DDR2 and not waiting for DDR3 as all the Rumors ran with?

8mb harddrives have been out for ages! (you're going back too far) Granted some crappy OEM systems came with only 2Mb drives a few years back.

I did say OEM's selling 2MB Cache HDDs, just as many are still selling 8MB now. One more time, my SATA2 16MB HDD kicks the crap out of my SATA1 8MB HDD.

But I was talking about CPU / ram / motherboard, anyway That the problem, they haven't been making much progress the past 2-3 years. We are in the processors forum, not the storage/hd forum.

Computers are systems, not one part or another. I AGREE there are times when progress seems like a gimmick. People in this market have been saying saying for at least 3 years, "Please get the rest of the tech to catch up with Processors, that means faster RAM DDR2+, Better HDDs SATA 3Gbs, better than AGP, PCI-E and etc.. Processors are overrated by everyone but Geeks in this market.

There are improvements that don't seem worth the time to upgrade but other things make you go wow! We're in a Processor forum but that Processor does NOT operate in a VOID. You can't devorce it from the rest of the system. I'm here bitchin' about motherboards when I know damned well I want something better than My IC7-G for a Conroe if I bought one. Yes Mobos do matter on the CPU thread. That's what caused AMD so much Pain and lost sales for so long. It's what causes few to still ask only for Intel. Example, would there be so much Hammer hpye if folks still had to install them in POS VIA motherboards like those early Kt266 and KX-133s?

I'm an Intel f4nboy, but you've got to admit their stock processors have made VERY little improvement over that past 2 years.

I completely skipped the 90nm process for Intel, b/c there was nothing worthwhile. Finally with Cedarmill and Presler there's a reason to upgrade from an old Northwood/875 system b/c you can really overclock the sh1t out of these new 65nm processors. (but if you're buying a stock system, you're getting about the same performance as something 2 years old. )

My current 3500+ is the first AMD system I've built for personal use in 3.5 years. I didn't like 90nm either BTW. My old 2.6C has served me very well. There's a lot here I agree with you about, just a few I don't. Left out of your performance equation are things like price per formance, badly needed Platform/Drive/Video improvements and Intel needed to fix terrible leakage problems for Prescott.. I already said many times that maybe Intel should have just made a smaller Northwood and added more cache to it while we wait for Conroe. I have NO doubt it would have reached 4 or 4.2GHz.
 
Northwood wasn't capable of running AMD64 (or EM64T, if you wish to call it that) instructions, that's part of why they made prescott. People would be bitching if Intel users had to wait till the middle of 2006 to do something that AMD users have been doing since mid 04.
 
robberbaron said:
Northwood wasn't capable of running AMD64 (or EM64T, if you wish to call it that) instructions, that's part of why they made prescott. People would be bitching if Intel users had to wait till the middle of 2006 to do something that AMD users have been doing since mid 04.

Are you sure they weren't capable of it?

Just like Hyperthreading was turned off at first... maybe Yamhill(EMT64) was just turned off.

It really doesn't matter as 64-bit support still sucks (OS, driver & app support) and its going to remain a blue-crystals feature (gimmick) for at least another year for 99% of people.
 
Back
Top