roast

Intel is planning to release a new Celeron chip - good to know they are'nt making LGA775 obsolete yet!

The new chip is to be released in 2010. Named the E3900, it will probably be the last model in the E3xxx series.
Its based on the wolfdale, has 1MB of L3 Cache and runs at a blistering 3.4GHZ But the best thing is, it is only 800Mhz QDR FSB.

Supports x64, but no intel VT or any trusted execution features. This is pretty much a basic workhorse chip.

I can imagine some great OC results from this. The low FSB/low cache combo always makes for a good OC. Not sure what the volts are, hopefully low with some nice headroom.

I'm definitly going to get one of these babys - comboed with a 630i platform, this would be a nice little benchtester for GFX cards and such..

Anyone have any opinions?

Phædrus2401

... A multiplier of 17!? I'd like to see a link, please.

roast

Sure! Theres even a CPU-Z screenie..



EDIT:1.2 voltage... not too high. Wonder what the rated maximum is?

-Mick.

Phædrus2401

Damn.
http://www.techpowerup.com/102396/In..._for_2010.html

That's like an overclocker's wet dream. And I thought having an x11 multi made me hot ****. Only thing is, with a multi that high it may be reaching the point where you derive no benefit from the extra clocks. You see, the multiplier is a ratio, of how many CPU cycles occur per bus cycle. In this case, that means that the CPU receives data from the bus--and then goes seventeen cycles without receiving more data! Commands that will take a full 17 clock cycles to execute are fairly rare--most are under 10 from what I understand. Thus you receive a set of commands to execute them, you finish them in say twelve clock cycles--then you're sitting around for another five waiting to send the answer back to the RAM and receive a new command. Thus 5/17 of your ultra-high clock speed is wasted in that instance.

So in this case, although having a low FSB with an extremely high multi means that you can overclock to ridiculous numbers (you can break the 4.0GHz barrier with just a 40MHz increase!), it may actually give less real world performance-per-clock-cycle than a CPU with a high FSB and low multi, like the E8x00 series.


EDIT: The voltage maximum for these 45nm chips is usually around 1.3625V. One thing I like about the E2xxx and E4xxx series, the 65nm architecture means you can take it as high as 1.5V before exceeding the rated maximum, and I've had this E2200 up to 1.56V without it frying.

roast

Very true.
In "synthetic" benchmarks, this chip would do very well.. but to be honest, I cant see any benefit in real world performance at all. From overclocking this, I think the only real benefit would be a huge increase in your e-*****... :P

I've spent a fair bit of time overclocking the E5200 - 2.5GHZ 800Mhz FSB with 1mb cache. Straight out of the box, I OC'd it from 800QDR FSB to 1066QDR FSB (not a huge overclock, I know) and got a result of 3.33GHz. Only raised temps by 4degrees on load.

If the E3900 performs as well as this, I could apply the same overclock and get 4.5GHz... thats pretty nice, and leaves plenty of headroom for more.

Just checked the spec sheet for the E7200 (45nm too) just before I saw this post. I like the way Intel decided such a low voltage as the max, when its capable of so much more. Good old Intel covering thier backs.

Mick.

Phædrus2401

You know, I wonder how this thing would perform if you dropped the multi to x12 and increased the FSB? Normally I don't like that form of overclocking because it puts more strain on the motherboard and increases instability. However, if this CPU is capable of such high clock speeds then there's no sense wasting them twiddling its thumbs waiting for instructions. A multi of x12 and FSB of 375 (1500... high, but manageable on high end boards) would give you a clock speed of 4500MHz. I expect that would be the most performance you could get out of this chip, and it's a lot of performance! Again, though, there's the issue of instability, and the bus wall. I don't know. Maybe a compromise? Multi = x14, FSB= 300MHz?

But though this is obviously appealing to overclockers who care more about clock speed than real performance, I think the real reason Intel has gone with this combination of low FSB and high multi is because it gives them the ability to put a processor with a very, very high clock speed onto very low-end motherboards that don't support the higher bus speed needed for CPUs like the E7x00 and E8x00 processors. Thus I expect to see a lot of these in 2010/11's OEM machines.

roast

Sounds tasty.
Judging by the trend I've seen on 45nm C2D's, I notice a general FSB wall of around 400 on average. Although, the E7200 I use doesnt hit a wall until 445-450MHz..
If this chip could do the same, it would be great.

I totally agree with your last statement.
Its also a lot cheaper for Intel to just lock the Multi at a different number instead of totally revising the core architecture just to allow a higher FSB.
As awesome as it seems to reach 4.5GHz on a cheap-*** chip, its just not going to cut it in "real world" performance, with a cache of 1mb...and I see this chip potentially bottlenecking high end cards, such as the upcoming 58xx or GTX300 series.

-Mick.

Phædrus2401

Still, if the price is <$140 I may go with it instead of the E8400 (I'm going to upgrade this time next year, probably). If it really is possible to drop the multi and raise the FSB as high as I would hope then this would make an excellent, excellent processor. However if it costs much more then I'd probably go with the E8400 just for better out-of-the-box real-life performance, plus the guarantee that I can get it to at least 3.8GHz (probably 4.0) and get actual performance gains out of it without having to worry about dropping the multi.

So it will be a neat experiment. I expect it will be a financial success because of the OEM market, so whether or not it flies with enthusiasts is moot to Intel. Let's just hope it turns out as nice as it looks it will.

greenbrucelee

I wonder if this will beat my 4GHz e8400, could be worth an investment if it can.

roast

Please let me know your evaluation of this CPU if you get it. I might get one myself.

Off topic - E8400 - very good choice. I shall own an E8600 in a few weeks. Second hand, from a good batch. Member of the "5.0GHz club" on another forum

Back on topic -
I'm sure you can drop the multi. On the E7200 and the E5200 I've tested, you could drop it on both down to a minimum of 6, which is a huge drop.
From that CPU-Z screenie above, it shows "6-17" beside the multi, so I'm assuming this is the range you can set it to in the BIOS.

Nice!

Mick.

EDIT:
Judging from the specs we've seen so far, it might. Although, we'll have to wait and see... knowing what stepping it would be based on would help.
I'm guessing the screenie above is from an engineering sample.

-Mick.

greenbrucelee

I reckon if the multi is dropped to 12 then I could get some fun out of it but I'll see how things pan out and hope that Asus make a BIOS revision for my rampage formula for it then I'll do some testing.

roast

Sounds good.

I heard that the rampage formula had a bit of an iffy BIOS... I have'nt used ASUS gear in a few years though, so I cant comment. A colleague says that he flashed the BIOS on his Rampage to the Maximus BIOS, and it was more stable. Perhaps you could have a look into that?

Thanks,
Mick.

greenbrucelee

I have never had an issue with it, I am running at revision 408 I think and havit at 3.91GHz at the moment but have had it higher.

The only issue there is with with the bios it does read the e8400 cpu temps incorrectly but thats a generall issue with the e8400 and the e8500 anyway they get stuck thermal sensors so using real temp is the only way to get accurate temps.

roast

Understood - was not aware of that issue with those chips.

Thanks!

Phædrus2401

Yeah, you can drop the multi to x6 on any Core 2 chip. The thing I'm worried about is if it will be stable at higher bus speeds. Even with my E2200's multi set at x8 (it gets a little unstable below that) I can't get the FSB over 360MHz. At the stock x11 I can get to 290 and still be stable, it will boot at 300MHz but runs very hot and crashes if you try to run a strenuous program like a game. I've found the best compromise between heat, stability, and performance is 270x11 @ 1.46V = 2970MHz.


What I would like to be able to reach with an E3900 would be 350x12 @ 1.35V = 4200MHz. I figure that would be the most you could get with good air cooling, assuming the chip runs really cool for its clock. With water cooling you could probably reach 375x12 @ 1.36V = 4500MHz which should be enough for anyone.