AMD or Intel as well as the memory brand - page 4

 
Svinozavr >> :

Of course it's written. Were you expecting Mickey Mouse there?))

Yeah... With Cheburashka and Geno in his arms...

)))

You don't understand these producers.

Sometimes they put important information in a prefix or a suffix,

or even something in the batch number like B0XN, and guess what...


Mothers. One description says s775, the other says s775 LGA

Just think... it's different or CM and FE are two people not four...

)))

 

In short, a netbook is best for trading. The main criterion is not speed, but battery life.

For optimisation and other stuff, the scarier the better.

 

Thanks for the tip on the i7 !!!

;)


RAM.

1 slab 1024mb 800mgts
Patriot 523r.
Kingston [KVR800D2N6/ 1G] 551r//[KVR800D2N5/ 1G] 619r.
Kingston HyperX 5-5-5-15 [KHX6400D2/ 1G] 604p.
Hynix 562p.
Samsung 567p.
NCP 533p.

This one doesn't have 1024, as a reference point:

Corsair // 2048\800 1369p.


Another manufacturer.

The price group is high...

OCZ
2048Mb PC2-6400 800MHz DDR2 DIMM OCZ Value [OCZ2V8002G] BOX 935p// order
2048Mb (2x1024Mb) PC2-6400 800MHz DIMM OCZ Flex EX XLC CL4 4-4-4-15 [OCZ2FXE800C42GK] BOX 1983r.

True, the spread here is quite significant.

 
Skymaster >> :

I've been reading some AMD and Intel tests here. So AMD Phenom II 965 beats the most powerful Intel i7 in most parameters. And it beat it in tests where purely mathematical calculations are used. Although there was one test where AMD simply tore Intel. That was game play. There AMD outperformed Intel by over 30%!!! At the same time the price tag of the most powerful AMD is the same as an average Intel.

Here's a link with tests of this top AMD stone. There is nothing like your statements, and even to the i7-920 this stone is a long way off. Quote from the final part:

As we've seen in the tests, the new Phenom II X4 965, running at 3.4GHz, performs about the same as the Core 2 Quad Q9550, rated at 2.83GHz, and lags behind the Core i7-920, which is even lower at 2.66GHz. So, AMD processors are quite seriously losing out to the competition in terms of IPC (number of instructions executed per clock). [...]

Also, considering that the Phenom II X4 965 has raised to 140W typical heat dissipation, its release is very much like a "last hope announcement".

Show your link to the tests please. Certainly AMD has closed a bit of a smashing gap with Intel, but Intel is not standing still either.

P.S. Oh, and by the way, the Phenom II X4 965 costs about 1700 roubles more than its direct competitor, the Core 2 Quad Q9550.

P.P.S. Another link - but there is another AMD stone, Phenom II X4 955 BE. Here the comparison shows even more clearly, who's the boss in the house.

 
Urain писал(а) >>

The need to reboot has a direct relation to the processor, when the CPU temperature increases, the number of errors increases, when the errors accumulate above a critical mass, the system crashes by itself or you reboot it, this effect does not depend on the computer.

It's not a CPU problem but a cooling system problem. My standard boxed cooler does its job successfully (even on a rather dusty heatsink).

 
Svinozavr >> :

Speaking of prices and application. For MT (one copy) - one flow. If you race test/optimisation, it's mostly frequency dependent. The i7 has the same compute core as the Pentium. There are no virtual threads in the new Pentium but it was the Pentium where they appeared for the first time! But then they were refused and came back to i7.

The price of the new 2-core Intel Pentium E6500K is under $90. And you can overclock it up like hell on a regular air cooler:

So think about it: do you need to overpay by a factor of 10 (even more!!!).

Voltage 1.7 is a lot for 45nm.

 
kombat >> :

No one is going to save money on matches, of course, but no one is going to overpay for anything either.

Making a choice based on the principle: better is more expensive

is not always justified, and at a price difference you can get the same thing just by overpaying for a sticker.

Brand, in terms of RAM and motherboards is important. For with the same components you get different products.

Both in quality of design (we are talking about motherboards), and (more importantly) in quality of build and quality of assembly equipment.

aha!

I wondered: why do they paint slots in different colours...

;)

By the way! Are there any ways of "protection from sucker" while buying - so they didn't push me a usual one instead of "extreme"?

If we talk about Socket 775 then Intel is better than AMD.

(Used computers do not have a brand name, only a residual value).

 
igar00 >> :

Voltage 1.7 is a lot for 45nm.

So what? A socket, for example, has even more voltage - and nothing. What did you mean by this post?

I'll let myself guess:

1) That it will fail? Successful overclocking is defined as stable operation in stress-tests with 100% stone load over a long period of time. Since I didn't overclock it, I can't swear on my mum. Overclockers claim the performance is stable.

2) CPU-Z does not correctly display corevoltage? Might be - again, I didn't mess around.

And that's not the point. And the point is that for MT purposes, all the i7 gimmicks aren't worth a damn. You can get similar performance with an MT for less than 90 quid, not 999. That's just the stones. And then there's the cost of the motherboards with their chipsets. Earth and sky.

 
It's not like you have your fingers in the socket all the time. You could burn out or degrade the stone. 1.7 was on Athlon 64. On Core2Duo 1.55 and more is useless (when the norm is 1.2). It only gets hotter and fails and heat dissipation and power consumption increase non-linearly. And the smaller the manufacturing process, the lower the threshold. 1.45 at the most according to Intel.
 

As a professional strength-testing engineer (Head of the Bureau of Strength Calculations) I officially declare: AMD processors are definitely better for mathematical calculations.

I have been using AMD CPUs at home since early '99. In terms of price/performance these devices were and are beyond comparison.

Reason: