SI
SI
discoversearch

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor.  We ask that you disable ad blocking while on Silicon Investor in the best interests of our community.  If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.
Politics : Formerly About Advanced Micro Devices

 Public ReplyPrvt ReplyMark as Last ReadFilePrevious 10Next 10PreviousNext  
To: Scumbria who wrote (54216)4/5/1999 12:24:00 PM
From: semi_infinite   Read Replies (2) of 1573590
 
Here's some interesting info from an ng...can't vouch for authenticity

<I work tech support at a computer wholeseller and enduser seller.
I test all things computer all damn day.

Intel smokes AMD is support and speed.
Les look at this one.

Microsoft makes the operating systems.
Microsoft supports intel processors more than AMD, so could we expect faster
computing straight out of setup froom AMD and win98? No. Need a stupid
patch to run anything in 3dNow accelerated registers. The difference
respectfully is:

AMD k6-3 333 running generic game loop demo no 3dnow technology:
say 30 frames per sec
ok, now with 3dnow:
say 31.5 frames per sec(wow huge difference)

Intel p2-333 running same hardware same company motherboard etc:
say 32 frames per sec
already better, and you can overclock that thing to push up the frames to
like 50, why would you want to use AMD?

I have tested the k5, k6, k6-2, k6-3, k7(early sample with motherboard),
cyrix 4x86, 5x86, 6x86, intel 8086, 80286, 80386, 80486, 80486x2, 80486x4,
80586, 80586mmx, celeron, p2, p3 oems

I know what the difference is. Intel's new p3's have 133 bus speed and
consequently the agp has a low(100) 2/3 speed, they're going for the high
speed as well as the support guys, if you can't figure that out, yer just
plain ignorant.

So what if AMD reaches a gigahert main platform pc before intel, it'll still
be lagging because of support. Ever bought a 3rd party manufactured sdram?
Like taiwan chips on a NEC ramboard? How well did yer computer like that?
It didn't. I couldn't really tell how fast the k7 system I tested was,
because I compared it to a p3-533(133x4.0) k7-600(200x3.0)

Tested all sorts of cpuload programs specially designed by both parties, and
the p3-533 was only 10mhz test below the k7-600 on both tests.

the damn k7 motherboard also drew too many watts for my power box to
handle(250 watt box). kept shutting down like 5 minutes into the test even
though the board was reletively cool.

Point is, intel is still the king and their r&d dept is not lacking in any
way, they're just sacrificing unused mhz in the name of support.

Don't assume cuz k7's have a 200bus that they will be faster than intel,
they're just overclocking their slot1 (100 on way in, 100 on way out, same
way agp2x does it)

So, after all things considered, yah the k7 outperforms the fastest p3 i
could get my hands on(barely). Ok the k7 is going to cost less on the mass
production, but when the p3-600+ hit the market, everyone is going to buy
them for one reason and it isn't a stupid marketing ploy. They are as fast
as you can get for the money, and everyone and their mother is supporting
them and not AMD.

Believe me I wish AMD sold processors, but as it is, even the k3-450s are a
poor man's shortcut to high megahertz. If people fell for prices alone, AMD
would be doing huge business, but who cares when ya can even overclock a
p3-450 to 633 with more volts/watts. (done it, kicks k7's ass all over)

Oh yah, l2-cache sucks, just thought I'd point that out, even though that
was a main-point of k7 architecture. Not only sucks, but sucks BIIIIG time.
It BARELY makes any difference at all unless they find some way to speed up
the nano's on them.

Gotta admit, the new copper processors are fast, good ol' copper.
Have fun AMDers, hope it sells, cuz I aint falling for price over support
again.
-John Doh! Es>
Report TOU ViolationShare This Post
 Public ReplyPrvt ReplyMark as Last ReadFilePrevious 10Next 10PreviousNext