> The new Intel i815 chipset uses PC133 SDRAM natively, for coppermine 
support
> up to 1GHz or whatever.  They benchmark similiarly to a 440BX based
> motherboard when both run at the same clock speed (i.e. both at 100Mhz, 
or
> both at 133MHz with the BX overclocked).

So that new component seems to be very interesting, especially moneywise 
for Intel, but not for the buyers.

> > This cheaper version of this Asus CUV4X MB uses the VIA 82C chips : I 
am
> > now very happy not having paid the extra 100 $ to buy the Asus Intel's
> > version, which as you say may have given the same result !
>
> Try the new Asus CUSL2.  From all reports, this is a winner.  Said i815
> chipset, uses PC133 SDRAM.  6 PCI slots, no ISA slots.  has AGP Pro 4X 
(the
> kind that actually works correctly in 4X mode with advanced graphics 
cards
> like the Geforce family), unlike the VIA stuff).  The chipset has a lame
> built in Intel graphics controller, this is probably adequate for 17"
> monitor business systems, but is easily disabled by simply installing a 
AGP
> card for higher resolutions or gaming.

OK, but I don't buy motherboards like bread (you know the "baguette" you 
eat with "camembert" here).

Thanks for the info, I'll check this one.

> Prime95 has a working set of about 5MB now when doing lucas lehmer tests 
on
> current range numbers.  These CPUs only have 256k cache. about 5 or 10 
times
> a second, the CPU is walking through the 4MB FFT arrays a NUMBER of 
times,
> including both reads *and* writes in rather complex patterns.  Its my
> understanding from what George has told us repeatedly that the LL tests
> positively HAMMER the memory bus to keep the FPU pipeline busy.

Mmmmh, I'm sure George no fool is man, but if my FPU works faster than my 
bus I would have appreciated less usage of the bus ...

> > George also talked about sdram speed, it may be the point too.

Replay to myselft, I've done some researches...

Yes memory was a major bottleneck in this case.
I checked the BIOS, having originaly let it set the SDRAM configuration at 
"Defined by SPD" or something.
In this configuration, it gave CAS latency, RAS precharge and RAS to CAS 
delay of 3T each.
In fact, I noticed this configuration corresponded to an 8ns SDRAM.
But mine is 7ns, so I input manually a config for the 7ns SDRAM, which 
makes only 2T at those delays.
Then Prime iteration went from 0.234 down to 0.219 (very approx 10 percent 
less).

I do conclude it's a kind of bugg in the BIOS because it should have 
detected the 7ns (only) 128M bank that is on the machine.
I may have some complaining to Asus if I get the time to.

Later, still being deceived not to acheive those 0.190 given in George's 
benchmark, I put up the speed from 100 to 112 Mhz, so now my first (and 
light) overclocked machine is going well to its 729 Mhz, with no special 
cooler.

Now iteration time is 0.196, i'm feeling kind of happier (life is stupid, 
no ?).

But because the weather becomes pretty hot here, i'll watch carefully to 
the cpu temp, so I won't be obliged to glue a sticker "Intel Outside" after 
being blowned out from the box by its own explosion.

Thanks also John for the object field correction.


Sylvain Perez

> > 
_________________________________________________________________________
> > Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm
> > Mersenne Prime FAQ      -- 
http://www.exu.ilstu.edu/mersenne/faq-mers.txt
>
> _________________________________________________________________________
> Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm
> Mersenne Prime FAQ      -- http://www.exu.ilstu.edu/mersenne/faq-mers.txt
_________________________________________________________________________
Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm
Mersenne Prime FAQ      -- http://www.exu.ilstu.edu/mersenne/faq-mers.txt

Reply via email to