When will the Northwood chips be released?

[Rick67]

Thanks,

 

[XWolfsraider]

last at week googlegear.com

retail $679.00 i believe for 2.2

 

[o1die]

Official release date Jan 3 or 7th. 1.8 and 1.6 versions available soon. If anyone needs a rambus mobo, newegg has the abit Th7-II with raid (refurbished with 30 day warranty) for only $135 with free shipping.

 

[MilkPowderR]

For $679, you could buy many many Athlon XPs each of them beating one $679 priced North-plywood, muhahahahahahahahaha. I am really laughing.

 

[apoppin]

<< For $679, you could buy many many Athlon XPs each of them beating one $679 priced North-plywood, muhahahahahahahahaha. I am really laughing. >>


I guess you missed that thread in General.

According to the benches, the 2.2Ghz Northwood is the fastest desktop processor (period)

It may certainly not win in price to performance but it leaves the fastest Athlon behind in raw peformance.

 

[AGodspeed]

It may certainly not win in price to performance but it leaves the fastest Athlon behind in raw peformance.

You're right, but NW leaves it " behind" by only 5% or so, and it's against only the 1900+. The Athlon XP 2000+ will be out the week NW is released, so...

Also, a Thoroughbred 2000+ (or 2200+) part will be released in March, so that's that it seems.

 

[apoppin]

In terms of Overclocking, it looks like the Northwood will also win. There are several reports of the 2.2Ghz o/c'd to 2.93Ghz with stock cooling. The 2000 XP+ won't go much beyond it's rated speed.

If you MUST have the fastest desktop processor - and price is not an object (as in use for work OR bragging rights) the Northwood is IT (now).

 

[ColdFlare]

the 2.9ghz could not bench as well as the 2.ghz xps. also when the .13 micron cheaps of xp come out the will also be as overclockable as the northwoods. lower micron, less voltage, more overclockability, unless intel locks the damn thing before it is released, multi and fsb.

 

[damocles]

<< the 2.9ghz could not bench as well as the 2.ghz xps. >>



Umm, I imagine a 2.9Ghz NW would trounce an Xp 2000+

Also, why are you so sure the .13um XPs will be as O/C friendly as the Northwppds?

 

[AGodspeed]

Umm, I imagine a 2.9Ghz NW would trounce an Xp 2000+

Also, why are you so sure the .13um XPs will be as O/C friendly as the Northwppds?

Notice he said 2GHz damocles.

Also, I remember reading somewhere that shrinking die sizes always helps a processor overclock better. Though I certainly could be wrong.

 

[christoph83]

For $679, you could buy many many Athlon XPs each of them beating one $679 priced North-plywood, muhahahahahahahahaha. I am really laughing.

Remember this CPU being sold at googlegear is before the actual launch date,expect that price to come down a good amount at launch and most likely more after a month or so. Not really fair to compare that price to anything since its not even officialy launched yet. Remember when the first P4's came out? They were priced almost near or above 1,000 dollars and came down considerably after a few weeks.

 

[ColdFlare]

<< Umm, I imagine a 2.9Ghz NW would trounce an Xp 2000+

Also, why are you so sure the .13um XPs will be as O/C friendly as the Northwppds?

Notice he said 2GHz damocles.

Also, I remember reading somewhere that shrinking die sizes always helps a processor overclock better. Though I certainly could be wrong. >>



yes in the past, die shrinks have always yeilded higher clock speeds, which also means higher overclocks. i think that makes sense!! hmm

also, the 1000 dollar pee4 that came out, went down to 800 dollars. the pee4 is selling for less now, but that is due to the price wars that amd and intel had.

 

[Degenerate]

Still surely a diff of 900mhz i think cannot be beaten. id say the two chips should equal at about 400mhz diff + - 100. Also, will the 2.0Ghz Xp be of the die shrink?

 

[AGodspeed]

Also, will the 2.0Ghz Xp be of the die shrink?

Absolutely. In fact, I bet we'll see the first .13 "die shrunk" Athlon at around 1.67-1.73 GHz.

A 2GHz .13 Athlon probably won't be around until sometime next summer.

 

[thermite88]

<< Also, why are you so sure the .13um XPs will be as O/C friendly as the Northwppds? >>

It won't.

The strategy of AMD to gain a hand in benchmark is to stuff their CPU with large L1 cache. When they do that, the CPU is not going to be very scalable nor overclocking friendly. L1 cache has to use fast static RAM which generate a lot of heat. There is not sign that AMD will change that direction, large L1.

It is still not sure how many months is AMD behind Intel in the 0.13 race.

 

[ColdFlare]

<<

<< Also, why are you so sure the .13um XPs will be as O/C friendly as the Northwppds? >>

It won't.

The strategy of AMD to gain a hand in benchmark is to stuff their CPU with large L1 cache. When they do that, the CPU is not going to be very scalable nor overclocking friendly. L1 cache has to use fast static RAM which generate a lot of heat. There is not sign that AMD will change that direction, large L1.

It is still not sure how many months is AMD behind Intel in the 0.13 race. >>



dude how can you say that it wont? the tbirds have large l1 cache and they are as over clocking friendly as ever. The tbirds right now are the hobbist chip which gives them the option of overclocking a lot. Remember the axia those oc atleast 600mhz alone with a large l1 cache, the XPs are ocing very much. I dont think its the L1 cache that is generating most of the heat, rather the structure of the chip. Amd has already converted their fabs up that are ready to produce .13 micron chips.

it is like amd to always let intel make the move first, before they release their ace.

 

[thermite88]

ColdFlare, go check the CPU database at overclockers.com. If you have another source, share it with us.

Static RAM generates a lot of heat. It is pure physics. No matter how many courses of electronics you take, you cannot fight mother nature.

 

[AGodspeed]

The strategy of AMD to gain a hand in benchmark is to stuff their CPU with large L1 cache. When they do that, the CPU is not going to be very scalable nor overclocking friendly. L1 cache has to use fast static RAM which generate a lot of heat. There is not sign that AMD will change that direction, large L1.

Using that exact same argument, you could infer that the .13 micron Celerons (Tualatins) don't overclock well. But of course, you would be greatly mistaken to think this, since it is a well known fact that the .13 micron Celerons (at, for example, 1.2GHz) can overclock to 1.5 and even 1.6GHz stabily on simple air cooling. Btw, if you didn't already know, the Tualatins have 128kb of L1 cache (which just shows that having a "large" L1 cache doesn't necessarily translate into poor overclocking).

Of course, the more cache you have, the "hotter" the die temperature will be. But that doesn't mean the .13 micron Athlon XP's won't overclock well when they debut next quarter (assuming they keep the same size 128kb L1 cache they have now, which is a safe bet).

Btw, aren't the new Northwood P4's going to have double the L2 cache they have now? Wouldn't this make for a "hotter" die and poorer overclocking for Northwood compared to current Willemette P4's? This contradicts recent news showing 2.2GHz Northwood's being overclocked to over 2.9GHz!

More on-die SRAM does NOT necessarily mean poorer overclocking results.

 

[thermite88]

Comparing putting 128KB L1 cache in a 0.13 die to 256KB L1 cache in a 0.18 die is comparing apple to orange. Or is it 32KB?

<< Btw, if you didn't already know, the Tualatins have 128kb of L1 cache >>

Oh, Why did Intel misled me all this time! All their spec said 32KB L1 only for the Tualatin Celeron.

Let us all hope that AMD will not try to maintain the same L1 cache ratio over Intel when they move over to 0.13 process. (Boy, I hate to imagine how hot the CPU may get if they keep 8 times more L1 than Intel, or 8x128=1MB L1.:Q)

Intel gets significant performance boost when they increase the cache size on 0.13 processors. It would be interesting to see what AMD may do to keep their "benchmark" edge over Intel.

 

[AGodspeed]

You are comparing puttting 128KB L1 in a 0.13 die to 256KB L1 in a 0.18 die (AMD). It is apple to orange.

First off, you have to get your numbers straight. The Athlon has exactly 128kb of L1 cache, not 256kb (it has 256kb of L2 cache though). The current .18 micron Athlon XP's have the same L1 and L2 cache sizes as the .13 micron Tualatin PIII architecture has.

Btw, if you do a search in General Hardware, you'll find that the .13 micron Thoroughbred Athlons will be no different architecturally (and that includes L1/L2 cache sizes) than the currently Athlon XPs according to AMD. This is very likely true, the .13 micron Athlons will have the same L1 and L2 cache sizes as the extremely overclocking friendly Celeron-II Tualatins have. Therefore, I doubt cache sizes will effect overclocking results in any significant way (again, just look at the Tualatins).

 

[thermite88]

<< The AMD Athlon&acirc;?¢ XP processor with performance-enhancing cache memory features 64K instruction and 64K data cache for a total of 128K L1 cache. 256K of integrated, on-chip L2 cache for a total of 384K full-speed, on-chip cache. >>

Directly from AMD.

AGodSpeed, you are so right. I apologize for my mistake.

Comparing putting 128KB L1 cache in a 0.13 die to 128KB L1 cache in a 0.18 die is comparing apple to orange. Or is it 32KB?

 

[AGodspeed]

Comparing putting 128KB L1 cache in a 0.13 die to 128KB L1 cache in a 0.18 die is comparing apple to orange. Or is it 32KB?

Yes, you are right, it's not exactly the same thing. However, since AMD has already publically stated that the Athlon will keep the same cache sizes going from .18 micron to .13 micron, we can assume that overclocking potential will increase.

Btw, I believe Intel hasn't updated their Celeron Cache Info page to include Tualatin cache sizes, since it states that the Celerons (I assume they mean EVERY Intel Celeron since they don't specify otherwise) have 128kb L2 caches. It's a known fact that the Tualatin Celerons have 256kb of L2 cache, and that the highest performing Tualatins (PIII) have 512kb of L2 cache.

This explains why you're thinking the Tualatin Celerons have 32kb of L1 cache.

 

[oldfart]

The Sspec page show the cache size.

Is there going to be a 1.6 and 1.8 Northwood? These would be nice overclockers.

 

[thermite88]

<< This explains why you're thinking the Tualatin Celerons have 32kb of L1 cache. >>

AGodSpeed, I apologize again. I don't have very good information source. I based mine on the Intel published Datasheet for Intel ® Celeron ® Processor for the PGA370 Socket at 1.20 GHz on 0.13 micron process in the US, not Australia. They (the US Intel) really did a terrible job mis-stating the Tualatin Celeron L1 cache size. Please, provide the AU datasheet link for the correct information.

<< The processor includes an integrated on-die, 256KB 8-way set associative level-two (L2) cache. The L2 cache implements the Advanced Transfer Cache Architecture with a 256-bit wide bus. The processor also includes a 16 KB level one (L1) instruction cache and 16 KB L1 data cache. These cache arrays run at the full speed of the processor core. The processor for the PGA370 socket has a dedicated L2 cache bus, thus maintaining the dual independent bus architecture to deliver high bus bandwidth and performance. Memory is cacheable for 64 GB of addressable memory space, allowing significant headroom for desktop systems. Refer to the Specification Update document for this processor to determine the cacheability and cache configuration options for a specific processor. Please contact your nearest Intel Sales Representative for the latest Processor Specification Update. >>



If you have problem with the Intel ftp site above, try this link.

 

[apoppin]

My .13 micron 866Mhz P-III Mobility (notebook) CPU has 32KB L1 cache (16KB data + 16KB instruction) and 512 KB L2 Cache.

My .13 micron 1.2Ghz (desktop) Celeron CPU has the same 32 KB of L1 cache and 256 KB L2 Cache.

This is reading from the Intel Utility.