• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

AM2 any real performance increase

i not entirely sure how dual core chips work, do they split the load between them at all, so one core is running mutliple background applications and the other core is running a single demanding application? anyways back to the argument at hand, with all the raving going on about conroe everyone has doomed AM2 before its even came into its own AM2 has one huge advantage over conroe which it cannot take advantage of because the timings on DDR2 are rubbish, its a well known fact K8 (if AM2 is still K8) relies massively on tight timings rather than bandwidth, where the intel chips with no onboard memory controller relied more heavily on bandwidth, i honestly thing everyone is making an enormous mistake in jumping on the conroe bandwagon before we have seen what AM2 can really do given low latency memory, in all benchmarks i have looked at AM2 still wipes the floor with conroe in memory latency tests and memory bandwidth tests, but the average i see is 44-ish ns in the latency tests for AM2 system, what the hell happened to the super low latencies of 939? easy to answer DDR400 with CAS2 latencies is much better for K8 than DDR2 with much slacker timings, i want to see fair tests, where one system isn't held back by something (namely AM2 being held back by rubbish timings on the memory) and im gonna assume that 'performance' gap between core duo and AM2 will drop significantly. also the end of this year when AMD release there 65nm AM2 chips, they'll clock fair bit higher than there current 90nm chips and hopefully by then DDR2 with better timings will be available so we can see conroe get the hell beat out of it for all to see. so IMO the answer to this threads question is definately a huge YES, but not until timings can be improved on DDR2 significantly, and who knows the 65nm chips might have a few performance tweeks as well, hopefully. personally im holding for AM2 to show its true potential
 
If you read the review I posted, some benchmarks showed Corsair DDR2-800 with CAS4 timings gave an improvement over DDR, more so than I expected.
 
yeah the chip makes use of the extra bandwidth as well as timings, still would be significantly faster if those timings came to CAS2
 
i would say thats entirely because of the on-die memory controller on the K8, its just more efficient than intels outdated system in every single way, plus look on the bright side, intel = evil :p and AMD have a better name, i mean come on...intel (boring) or advanced micro devices, might just be me but it sounds cooler :D
 
Gashman said:
i would say thats entirely because of the on-die memory controller on the K8, its just more efficient than intels outdated system in every single way, plus look on the bright side, intel = evil :p and AMD have a better name, i mean come on...intel (boring) or advanced micro devices, might just be me but it sounds cooler :D
Click to expand...
:confused:

There's nothing inherently "outdated" in the fact that Intel doesn't use on-die memory controllers.
 
Gashman said:
i not entirely sure how dual core chips work, do they split the load between them at all, so one core is running mutliple background applications and the other core is running a single demanding application?
Click to expand...
Dual core chips don't 'do' anything. It simply presents 2 cores to the windows task scheduler rather than one.
It is windows that decides what to run on which core, and it will intelligently allocate threads to cores.
Gashman said:
i would say thats entirely because of the on-die memory controller on the K8, its just more efficient than intels outdated system in every single way, plus look on the bright side, intel = evil :p and AMD have a better name, i mean come on...intel (boring) or advanced micro devices, might just be me but it sounds cooler :D
Click to expand...
So you would rather there was no competition at all in the cpu market and just have AMD charging whatever they like for whatever they like, just like windows?
You do realise that AMD's on-die memory controller and most other technogies exist because of competition from Intel, and if Intel didn't exist they would probably never have moved on from the fsb. Technological progress would grind to a halt, just as it mostly has in windows.
 
Last edited:
Joe42 said:
Technological progress would grind to a halt, just as it mostly has in windows.
Click to expand...
:confused:??? Heard of Vista? DirectX 10? Windows Presentation Framework ("Avalon")? PowerShell ("Monad")? Windows Communication Foundation ("Indigo")? WinFS? .NET?
 
NathanE said:
:confused:??? Heard of Vista? DirectX 10? Windows Presentation Framework ("Avalon")? PowerShell ("Monad")? Windows Communication Foundation ("Indigo")? WinFS?
Click to expand...
Certainly Have.
It'll be out, wait, 5 years after XP from a company with more money and resources than many small countries.
Half the features such as WinFS have been ditched, and the rest of them are copied directly from OSX anyway.

And will it be stable? Not a chance in hell.

Its not as good as it would be if it had a true competitor.

Can you honestly say it wouldn't be better if it was competing directly with another mainstream OS?
 
Joe42 said:
Certainly Have.
It'll be out, wait, 5 years after XP from a company with more money and resources than many small countries.
Half the features such as WinFS have been ditched, and the rest of them are copied directly from OSX anyway.

And will it be stable? Not a chance in hell.

Its not as good as it would be if it had a true competitor.

Can you honestly say it wouldn't be better if it was competing directly with another mainstream OS?
Click to expand...
Ignorance is bliss ;)
WinFS is almost to the Beta 2 stage and will be added to Vista in a later service pack.

Almost all the key features of Vista (such as all the ones I mentioned) predate the OS X operating system entirely. It just so happens Apple is a small fast and agile company that could bring them to market quicker.

Stability... well people used to say XP SP2 was unstable... before that people used to say XP RTM was unstable... and before that they used to say 2000 was unstable... it never ends. As it happens Vista will be the most reliable Windows OS yet due to it shifting drivers out of the kernel mode and back into the user mode. Will there still be buggy drivers for the first few months? Sure. Will there still be bugs that need working out for SP1? Sure. Will Vista be regarded as the most stable Microsoft OS to date when hindsight sets in? Yes.

Obviously if there was another key player nibbling at Windows' market share then it would be even better. But to say that technological progress/innovation etc has ground to a halt in Windows is a complete and utter load of rubbish, frankly.
 
Last edited:
NathanE: I agree with you. it has to be remember that apple designed OS's and software for their own custum designed platform, m$ has to cater for a nearly infinite number of different system configurations. m$ may be a much larger company than apple but they also have to look after many more markets than apple, OS are just one of their many markets so this huge amount of money they have for R&D is split between things such as console, MSN. servers and workstations OS, programming languages, Office applications, Windows Mobile and many more
 
paddy001 said:
Couldnt agree more, I heard just recently that a couple of chinese companies were begining to make good processors. Competition in the market is always best for consumers look at the gfx performance leaps over the past 12 months.
Click to expand...

I am chinese so dont think I am chinese bashing. Its sad but:

The chinese head of behind those nice chips has been fired and is probably going to be arrested. Faked results and it could not play mp3's and decode finger prints or whatever they claimed. Turns out the core itself wasnt even made by that company.

http://www.reghardware.co.uk/2006/05/15/chinese_chip_fake_claims/
http://www.theinquirer.net/Default.aspx?article=31680

Same kinda thing with the korean scientists and his cloning fake results
 
You must log in or register to reply here.
Back
Top Bottom