Low Latency, the lie...

Status
Not open for further replies.
ChaosBlizzard said:

"Remember your HAL controls threads, and everything has threads, your OS, your apps, everything. So lets say you bust open Quake 2, which is single threaded. Will you see a performance increase with two CPUs over one? YES. Why? The HAL can dedicate a CPU to the Quake 2 thread, and run the OSs threads through the Other CPU. The Machine will run better. If your run a multi-threaded app, the HAL can grab the threads from that program and execute two threads at once. It will assign one to each CPU. This brings up a key point. If a p2 450 can only run one thread, and a dual p2 450 can run two, then why is there not a 100% performance increase? The two CPUs must talk to each other. This bus overhead causes a drop in the performance of the CPUÂ’s. So, in the end, you get an 80% increase."


Do you understand yet?!?!

i understand one thing, you are wrong...

do not feel bad 4w4k3, you belong in this forum, people who refuse to acknowledge the facts and claim to know so much who cant even say that their system will be outperformed ever, dont belong in this forum...

example #1
example #2
"An abstract analysis looks like this: more is not less - but it is not always better. Anyone who wants one of the boards that we tested for switching over to two CPUs later should save their money instead. Even the argument that dual boards can create a powerful graphics or CAD workstation is not very convincing.

In the end, even professional CAD/CAE/CAM applications do not permanently access two CPUs. The reason is that multiprocessor support is a far cry from consistent, balanced utilization of both CPUs.
"
example #3
"clearly shows that dual operation does not necessarily mean increased speed for all applications. Rather, you need software that has been specially adapted to multiprocessor operation, so that the load is equally distributed between the two CPUs"

its bad enough you blocked PM's cause you were losing our last argument but god,

oh well, this is once again a pointless thread much like the last one we argued in chaos, perhaps you cant read thread titles or find yourself incapable of staying on topic either way please keep your posting in threads on topic, you can argue comps all you want, but this is bad procedure to run a thread about latency times into a dual processor vs single processor fight.

and BTW, your comp sucks :p :p

learn to take a GD joke

EDIT: oh yah, forgot one thing, could you please stop insulting all the other people in the forum for their intelligence... we are all hoping you will leave, not everyone else on these forums :)
 
I am wrong? That quoted text was taken from a bench test website from someone who knows a lot more than you. Also, Scott Muller seems to agree with how SMP works according to that quoted text. So your saying he is wrong. Scott Muller is the most Authoritive hardware voice in the industry. Go email him and tell him that statement is wrong.

More is not better? Perhaps in the world of debt.. But when we talk about CPU's, being able to do more IPC is always better. I can't help it your acting foolish.

My system process manager could tell you a different story about HAL balance. Your clams are unfounded. Not only do you not have an SMP machine, but you quote text from an irrelevant comparison.

Tom would be incorrect in his analysis of SMP. He's nothing more than a bench tester.. You don't see people quoting him in magazines, you don't see him on Tech-TV. However Scott Muller has been on Tech-TV numerous times.

I never said my system can't be outperformed. I said it could beat a slower less efficient CPU/System. Which happens to be many systems. You are just ticked because you pulled your computer out of a cereal box.

If people want me to leave, that is simple because they would rather not have to hear facts or listen to common sense. They would rather stay in their own little world of misconception. Your the only person here who can't take an arugement.

BTW- If you did your homework on Super_pi, you would realize it doesn't really use Floating Point Integer's to calculate any of it's math. Next time ask me to use a competent bench application.

Now since all your references are from tomshardware, this means you must take any information off of any site and deem it the master answer. Take your own advice and dig around.

I never blocked your PM's... I can't help it you can't find the send PM button.

Buy a book and stop complaining just because you don't know:
http://www.amazon.com/exec/obidos/t...104-3321249-5821559?v=glance&s=books&n=507846

slvrstang said:
Why are you so afraid (blizzard) to just try Super_pi. It is a very popular CPU/Ram benchmark. Its very small and completes in under 1min (on decent systems). I am curious as to what both of your scores would be. I think my best was 41secs on the 1M test.

I did try super_pi, the program stopped responding. I went on the internet to find a newer release. Instead I found about 50 people who use AMD chips saying the program doesn't really use FP calculations, and is therefore irrelevant. My system of course was unaffected by the cheap application. That was by the way written in the mid 90's...
 
Chaos blizzard...for someone who goes to a fairly well respected college you have very little respect towards others and open mindedness...it seems that 'nerds' feel all powerful against one another behind the computer screen. The fact is you aren't completely correct and if you really want to back up some claims and answer questions you'd try Super_Pi, I've heard of it used many a times in OC'ing forums...

Also your memory is performing horribly technically, mine gets 2500mb/sec compared to your 1000mb/sec or less when referring to the chart YOU linked.

Beyond that, this is the 2nd time you can't have an informative debate but more of a ranting arguement on your part and a lot of "No I'm right I'm going to ITT-Tech so I know everything" when that's not at all true.....maybe if you were a professor then I'd listen, but since you are a learning student then I'd say you need to check yourself and get this 'Superiority complex' out of here

Your only defense was 'You FORCED your CPU to run faster'.....yeah and so? Force yours faster and I'm sure his and mine both will still beat your Arithmatic...forcing the CPU to run faster is part of the game, you can't disqualify it for that reason, beyond that there are chips that'll run faster speeds than our stock chips, but we choose to save money and get the most bang for the buck.

I'm closing this thread because it has been way past Hijacked at this point. Leonidas I hope you don't mind, feel free to PM me if you wish. Other than that, no one else really has a say in it since Leonidas is the thread creator.

BE KIND OR SHUT UP

- Nubius
 
Status
Not open for further replies.
Back
Top Bottom