Pentium M beating P4 & FX?...????

[Scali]

The Pentium 4 @ 1.4GHz was handily thrashed by Tualatin Pentium III's at 1.13GHz.

Even Willamettes went all the way up to 2.0 GHz... The 1.4 GHz was the slowest, and that Tualatin wasn't even released yet, I believe. But the faster P4s were out of reach of the P3 pretty much from day 1.

Really until the Athlon 64 came out I've never been impressed with AMD. I must say the Opteron platform as a whole impresses me even more. The way they''ve designed scalability into the whole platform is amazing. Wheras Intel included SMP as more of an after thought because higher end machines had more CPU's and there is a market for that in x86.

Intel's REAL SMP system is Itanium ofcourse. You should compare the CPU that they specifically designed for SMP against AMD's one. The Xeon was always artificially limited, so that it would remain out of the way of the Itanium, both in price and performance. Intel considered Xeon as SMP-on-a-budget, while for AMD Opteron is top-of-the-line. But AMD has now awoken the giant, and I think they'll be very sorry they did

 

[chrisf6969]

Finally, a nice mobo for Pentium Ms, too bad its priced insanely high!!!
http://www.newegg.com/Product/Product.asp?Item=N82E16813137061

If it was about $100 cheaper, I'd buy it!!

Any Guinea Pigs out there wanna buy it and overclock a P-M with a high end videocard etc? I'd bet you get some impressive #'s.

I like that it still uses DDR. Makes me want to get one of these, so I can re use my old ram.

 

[computerpro3]

http://www.tomshardware.com/cpu/20050525/pentium4-10.html

I was pretty shocked when I read this and I almost find it impossible to believe. Apparently ASUS came out with a kit that allowed you to put a Pentium M processor into your desktop. At first I was like, "yeah yeah the Pent M is mobile and peforms like crap," but apparently putting it in a board with a decent chipset and memory makes one helluva difference! They show this baby beating the almighty Athlon 64 FX series and completely trashing the P4 series. Not only that, but it produces an EXTREMELY low amount of heat since it was designed for notebooks. Watercool this sucker, and you could achieve some INSANE overclocks.


Intel.... wtf were you thinking. Your could have owned the desktop market with this techlology.


Comments?

My friend just got one and at stock voltage for his 1.6ghz dothan hes rock stable @ 2.4ghz and cranking out a 31sec superpi without any optimizations.

That soundly beats my a64 3500 @ 2.6ghz which is max I could do on air.

 

[Dan_D]

Even Willamettes went all the way up to 2.0 GHz... The 1.4 GHz was the slowest, and that Tualatin wasn't even released yet, I believe. But the faster P4s were out of reach of the P3 pretty much from day 1.



Intel's REAL SMP system is Itanium ofcourse. You should compare the CPU that they specifically designed for SMP against AMD's one. The Xeon was always artificially limited, so that it would remain out of the way of the Itanium, both in price and performance. Intel considered Xeon as SMP-on-a-budget, while for AMD Opteron is top-of-the-line. But AMD has now awoken the giant, and I think they'll be very sorry they did

The 1.8GHz Pentium 4 was probably the fastest out when the Tualatin was released. However the Tualatin kicked the P4's ass all the way up to almost 1.8GHz. Of course it did come later. I do understand the point you are making. So while the P4 was a step in the right direction in terms of "performance". However the earlier P4 architecture was worse performing per clock than the previous processor architecture.

The same thing happened between the Pentium II and Pentium Pro. The Pentium Pro was faster at 32 bit code, of course that is primarily due to the cache being on-die and there was also version of the processor that had significantly more cache than the P2. All on die. When clocked at the same speeds, the PPro was always faster. Of course the PPro's didn't do 16bit code for crap and weren't stable with Windows 9x, due to the hybrid 16/32 bit code it contained. The Pentium II was an improvement there. Plus the PII was cheaper to produce.

So the Pentium II had numerous advantages over the Pentium Pro, while being "slightly" worse at the same speed. Those were advantages that counted. The Pentium 4 wasn't really. Only by virtue of being clocked insanely high compared to every other processor on the market did it win anything. However the Athlon XP's bested it at nearly everything until the Northwood came out.

I don't think its fair to compare Itanium to the Opteron. For one reason, Itanium doesn't run x86 code natively. That puts them in slightly different classes. Is the x86-64 and EM64T not very different than the IA64 code that Itanium executes?

 

[Big Worm]

The 1.8GHz Pentium 4 was probably the fastest out when the Tualatin was released. However the Tualatin kicked the P4's ass all the way up to almost 1.8GHz. Of course it did come later. I do understand the point you are making. So while the P4 was a step in the right direction in terms of "performance". However the earlier P4 architecture was worse performing per clock than the previous processor architecture.

The same thing happened between the Pentium II and Pentium Pro. The Pentium Pro was faster at 32 bit code, of course that is primarily due to the cache being on-die and there was also version of the processor that had significantly more cache than the P2. All on die. When clocked at the same speeds, the PPro was always faster. Of course the PPro's didn't do 16bit code for crap and weren't stable with Windows 9x, due to the hybrid 16/32 bit code it contained. The Pentium II was an improvement there. Plus the PII was cheaper to produce.

So the Pentium II had numerous advantages over the Pentium Pro, while being "slightly" worse at the same speed. Those were advantages that counted. The Pentium 4 wasn't really. Only by virtue of being clocked insanely high compared to every other processor on the market did it win anything. However the Athlon XP's bested it at nearly everything until the Northwood came out.

I don't think its fair to compare Itanium to the Opteron. For one reason, Itanium doesn't run x86 code natively. That puts them in slightly different classes. Is the x86-64 and EM64T not very different than the IA64 code that Itanium executes?

Both the Tualatin vs P4 and the Pro vs P2 does not count.

I believe he was refering to the fact that Intel has not release cpu's in the same class at lower performance then its predecessor. The Tualatin CPU was not intended for Desktop use. It was more of a middle of the road server CPU. as well the PPro was a full server CPU.

As well I *think* overall the first P4's were a tad faster then the 1.0Ghz P3.

 

[Scali]

So while the P4 was a step in the right direction in terms of "performance". However the earlier P4 architecture was worse performing per clock than the previous processor architecture.

Ah, the variation on the MHz myth, the IPC myth... When are we ever going to debunk that one?
It doesn't matter that the P4 performed worse per clk, because it more than compensated with its clockspeed.

The same thing happened between the Pentium II and Pentium Pro. The Pentium Pro was faster at 32 bit code, of course that is primarily due to the cache being on-die and there was also version of the processor that had significantly more cache than the P2. All on die. When clocked at the same speeds, the PPro was always faster. Of course the PPro's didn't do 16bit code for crap and weren't stable with Windows 9x, due to the hybrid 16/32 bit code it contained. The Pentium II was an improvement there. Plus the PII was cheaper to produce.

The fastest ever PPro was 200 MHz, while the slowest ever PII was 233 MHz. So again, Intel offered more actual real-world performance, despite perhaps a minor decrease in IPC.

Only by virtue of being clocked insanely high compared to every other processor on the market did it win anything.

The P4 had SSE2, which was an advantage aswell, and ofcourse its superfast FSB and Rambus memory. In certain areas its performance was unmatched.
Besides, what's wrong with 'being clocked insanely high'? The CPUs were still priced competitively. You paid a lot less per MHz than with the Athlon or the P3. Good deal, if you ask me. In fact, that's why AMD had to start its performance rating... It had to justify its low clockspeeds at high prices in some way to the customer.

I don't think its fair to compare Itanium to the Opteron. For one reason, Itanium doesn't run x86 code natively. That puts them in slightly different classes. Is the x86-64 and EM64T not very different than the IA64 code that Itanium executes?

What does that matter? A large part of the server market is based on *NIX systems, where the architecture is pretty much irrelevant, since most software is portable. You can just pick the architecture that performs best for you. Which makes IA64 an advantage, not a disadvantage over EM64T.
Besides, weren't we discussing who had the best SMP solutions? I don't see how SMP has to be x86 in order to 'be fair'.

 

[BossNoodleKaboodle]

Besides, weren't we discussing who had the best SMP solutions? I don't see how SMP has to be x86 in order to 'be fair'.

IBM destroys them all at it. 16 logical processors, 36MB of 1/2speed cache all on one piece of ceramic and silicon.

 

[BossNoodleKaboodle]

Finally, a nice mobo for Pentium Ms, too bad its priced insanely high!!!
http://www.newegg.com/Product/Product.asp?Item=N82E16813137061

If it was about $100 cheaper, I'd buy it!!

Any Guinea Pigs out there wanna buy it and overclock a P-M with a high end videocard etc? I'd bet you get some impressive #'s.

I like that it still uses DDR. Makes me want to get one of these, so I can re use my old ram.

I'm putting together a Dothan 1.7/512MB/6200-->6600GT from newegg.

 

[Scali]

Besides, weren't we discussing who had the best SMP solutions? I don't see how SMP has to be x86 in order to 'be fair'.

IBM destroys them all at it. 16 logical processors, 36MB of 1/2speed cache all on one piece of ceramic and silicon.

Well yes, but the claim was that AMD had better SMP solutions than Intel. IBM was never mentioned.

 

[SLee]

Really until the Athlon 64 came out I've never been impressed with AMD. I must say the Opteron platform as a whole impresses me even more. The way they''ve designed scalability into the whole platform is amazing. Wheras Intel included SMP as more of an after thought because higher end machines had more CPU's and there is a market for that in x86.

The scalability of the Opteron is good, especially for certain fields but it shouldn't be overrated. Opterons don't scale all that well past 4 sockets gluelessly with typical business applications, while Xeons used with a top-notch chipset like IBM's X3 can match or exceed the Opteron. In tpc-c, the highly influential database benchmark, IBM's x366 using 4 3.66 Cranford Xeons scores 141K-150K tpmC vs 130K for the 4 way HP DL585 with 2.6GHz Opterons. The 8 way x460 with 3.33GHz Xeons/8MBL3 scores 250K vs the 187K score for the 8 core/4 socket DL585-G1 with dual core 2.2GHz Opterons.