AMD 1700+ > AMD 64 3000 939

[kamaboko]

might be a stupid question, but the AMD 1700 runs at 1.47GHz and the AMD 62 3000 939 at 1.8GHz. of course the cache size is different, but should i expect double the performance speed? if not, what might i expect in percentage increase?

 

[imported_omega1]

there will be a HUGE difference, the performance rating makes an amd chip run about as fast as a p4 of the rating amd gives... although they SAY its based on one of their older cores LOL

 

[Cerb]

More cache, faster cache, on-die memory controller (faster RAM access), better branch prediction, lots of stuff AMD won't say, and no narrow FSB to get clogged.

 

[StrangerGuy]

The increase in CPU preformance will be like 2x if you upgraded to a A64 3000+ from a XP1700+.

 

[crisscross]

I would like to know too... faster in what aspects? boot times, application load times, game frame rates... what?

 

[Kccdx2]

everything.

 

[CraigRT]

It depends on what you do..... but even just chilling in Windows you will proably notice a difference with day to day tasks.

Fire up a 3D game and let the A64 open up..... Then you will see a huge difference.

 

[biostud]

AMD 1700+ < AMD 64 3000 939



And definately a lot faster for gaming.

 

[Cerb]

Take an already efficient chip, enter 5 years or so of real tweaks (per-clock performance from slot Athlons to Bartons have been entirely dependent of cache).
Cache that is faster by an order of magnitude.
Better register renaming.
Better branch predicting.
Remove a crippling 3.2GB/s FSB for a 3.2GB/s each way HT link, with RAM having its own bus to get to the CPU, no sharing.
Remove the latency of going to the chipset, then out to RAM. Speed up the RAM controller on top of that.
Then there are tweaks I haven't mentioned, and many which AMD won't say a word about (have you seen how many patents they made? And that was just for the x86-64...who knows how many "trade secrets" they have?).
All this without sacrificing a design with short pipelines, so it is great for branching tasks and office multitasking.

There are a handful of synthetic math benches where the A64 is only 10% or so faster than an equivalently-clocked Athlon, so many thinks are similar...but in EVERY real benchmark (as in how real software perform doing real tasks), the A64 handily beats Athlons clocked significantly higher.

As far as boot times, that's mostly HDD limited. A SCSI HDD will get faster boot times .

 

[magomago]

I wouldn't expect outright 2x preformance~ jumping from a 1.7Ghz P4 to a 3GhzP4 also probably don't yield 2x the preformance...but i would espect to see atlesat 50% faster computer in most situations, and if you gotta good videocard, maybe even 2x when you play games since th eA64 does that very well

 

[Amaroque]

Originally posted by: crisscross
I would like to know too... faster in what aspects? boot times, application load times, game frame rates... what?

It's more than 2x as fast as a Barton (at the same speed, 2.2GHz) when crunching prime numbers.

 

[xsilver]

Not to be an AMD hater but has anyone switched from a P4 with HT to and AMD 64 and noticed some wierd stuff going on when multitasking?
I used to surf, download, and watch a dvd at the same time fine but now the dvd stutters a lot -- is it maybe cool'n'quiet acting too slow to keep up with the CPU usage?

But to the OP since you're upgrading from an AXP1700 you will find a big difference, epecially if you upgrade the ram too

 

[bobalong]

I beleive the 1.7 runs at 133mhz with only 256L2 cashe with a PR rating of around 1800.

Compare that to the A64 with a PR rating of five figures - this depends slightly on the core and cashe size.

The PR rating is equivilent (if I remeber rightly) to the original P2's.

My friends P4 HT(2.8 clocked to 3.4) burned out after a PSU voltage spike, he replaced it with an AMD clawhammer 3.2. and noticed a slow down when running several applications similtanously, but made up the difference during gaming, boot times and stability

 

[Cerb]

xsilver: I don't know about a A64, but I'd have to say that's probably drivers (if using VIA)...else something wierd. You should be able to do all of that just fine, with no stuttering, without a P4 or HT...my AXP can play a DVD w/o stuttering while I do other crap.

I have noticed in some more recent PCs that P4s seem less sluggish with a bunch of background crap, though. With a typical set of OEM crap loaded, A64 desktop machines feel like snails. Have any websites done the business Winstone on Dell and HP boxes vs. nice PCs that aren't loaded down? It'd be really cool to have factual evidence of this. My own use has always matched the business winstone tests...it'd be interesting to see if it is the same w/ OEM boxes and their mile-long system trays .

 

[imported_Computer MAn]

Originally posted by: Cerb
Take an already efficient chip, enter 5 years or so of real tweaks (per-clock performance from slot Athlons to Bartons have been entirely dependent of cache).
Cache that is faster by an order of magnitude.
Better register renaming.
Better branch predicting.
Remove a crippling 3.2GB/s FSB for a 3.2GB/s each way HT link, with RAM having its own bus to get to the CPU, no sharing.
Remove the latency of going to the chipset, then out to RAM. Speed up the RAM controller on top of that.
Then there are tweaks I haven't mentioned, and many which AMD won't say a word about (have you seen how many patents they made? And that was just for the x86-64...who knows how many "trade secrets" they have?).
All this without sacrificing a design with short pipelines, so it is great for branching tasks and office multitasking.

There are a handful of synthetic math benches where the A64 is only 10% or so faster than an equivalently-clocked Athlon, so many thinks are similar...but in EVERY real benchmark (as in how real software perform doing real tasks), the A64 handily beats Athlons clocked significantly higher.

As far as boot times, that's mostly HDD limited. A SCSI HDD will get faster boot times .