K8MAN
Distinguished
- Apr 1, 2005
- 839
- 0
- 18,980
[LowBlow]
Well he could have gone to the store to buy some more mag's to look for more buzzwords and such to discuss to make himself feel like he actually has the slightest bit of knowledge of computers and the computer industry.
[/LowBlow]
The know-most-of-it-all formally known as BOBSHACK
Well he could have gone to the store to buy some more mag's to look for more buzzwords and such to discuss to make himself feel like he actually has the slightest bit of knowledge of computers and the computer industry.
[/LowBlow]
The know-most-of-it-all formally known as BOBSHACK
Atolsammeek
Distinguished
- Dec 31, 2007
- 1,112
- 0
- 19,280
Well I will not buy a dual core cpu. Untel there more uses for them. Right now if I needed a system I would go for a single core cpu that had 64 bit. Just due to the fact Most New cores are 64 bit.
1. I will never buy the newest chip. Meaning dual core.
2. I will never buy the frist genragtion of OS.
My rule on computers is this. Upgrade every 3, to 4 years. Laptops 4 to 5 years.
1. I will never buy the newest chip. Meaning dual core.
2. I will never buy the frist genragtion of OS.
My rule on computers is this. Upgrade every 3, to 4 years. Laptops 4 to 5 years.
>Dual core is good idea, but again pretty useless for 95% of
>the applications today,
Yep and for 90% of tomorrows (desktop) applications too. But then the same has always been true for every new technology going mainstream, like floating point coprocessors one day or SIMD instructions not so long ago. The big difference now is that the potential is bigger, but at the same time you pay a stiff price for it (lower clock and twice the diesize).
>It has been years and the performance improvements of CPU
>has really just stalled and no sign of any improvements in
>the future.
They keep improving, and will keep on improving. Yes it has slowed down a bit the last year or so, mostly because of netburst hitting a thermal brick wall and AMD not feeling the need to push their designs by lack of competition in the high end. If you'd chart CPU performance over time, youd see the improvements in the 1999-2001 timeframe where extraordinary (mostly due to AMD leapfrogging intel for the first time), but in general the current rate is not abnormal.
>The industry used to provide a 2X improvement in
>performance each year, now we see 10% improvements each
>year.
It never did 100% per year sustained; what has changed is that instead of large jumps every year (or even few years) we now see many small increases. For instance, the original Pentium was launched in early 1993 at 60/66 MHz. It took more than a year for a new one to hit the market at 90 Mhz, then another year to hit 120 Mhz and for 166 MHz you had to wait until a full three years (till 1996) from the original 66 MHz one. Now we see 5-10% improvements every so many months.
>What happened to the "we can go to 8Ghz with the 90nm
>process..." yada yada yada.
It referred to netburst (but I doubt the process node was mentioned), and netburst hit a thermal brick wall.
>And the motherboards are just as guilty of the lack of pure
> performance improvemen
Motherboard makers just implement, they never make the leaps, they cant even if they could. It takes chipsets and cpu's made by Intel/AMD/nVidia/via,.. but yes, FSB dont scale as fast cpus, and neither does the RAM. That is nothing new though; in the 486 (pre DX2) era CPU, FSB and RAM all ran synchronous. Ever since the cpu has outpaced the other 2 and that wont change ever. You could either blame FSB/RAM tech, or credit cpu improvements, but its not without reason AMD got rid of the FSB bottleneck, and likewise I expect embedded RAM to become the next step. in fact, large L2/3 caches could be considered just that already.
>The only tech industry that shows signs of improvement
>comes from Graphics processors (nVidia and ATI).
really ? All they do is make their GPUs wider with each respin, its very much like making multicore CPUs, except that 3D apps can actually make good use of them. Furthermore its about time nVidia and ATI start taking power consumption seriously. I read elsewhere that current high end videocards consume up to 50W *iddle*, there is no excuse for that.
>Is the mass migration of jobs to bio-tech a true indication
> of the state of the tech industry?
LOL, no. It is true however that the market is being commodotized more and more. just like we dont care much anymore about the "performance" of our microwave ovens, the number of ppl that need/want ever faster performance is shrinking rapidly. But we are certainly not anywhere near the end of technology scaling yet.
>Have AMD/Intel out sourced so much so that innovation is no
> longer possible?
They outsource next to nothing, and this has absolutely nothing to do with anything.
= The views stated herein are my personal views, and not necessarily the views of my wife. =
>the applications today,
Yep and for 90% of tomorrows (desktop) applications too. But then the same has always been true for every new technology going mainstream, like floating point coprocessors one day or SIMD instructions not so long ago. The big difference now is that the potential is bigger, but at the same time you pay a stiff price for it (lower clock and twice the diesize).
>It has been years and the performance improvements of CPU
>has really just stalled and no sign of any improvements in
>the future.
They keep improving, and will keep on improving. Yes it has slowed down a bit the last year or so, mostly because of netburst hitting a thermal brick wall and AMD not feeling the need to push their designs by lack of competition in the high end. If you'd chart CPU performance over time, youd see the improvements in the 1999-2001 timeframe where extraordinary (mostly due to AMD leapfrogging intel for the first time), but in general the current rate is not abnormal.
>The industry used to provide a 2X improvement in
>performance each year, now we see 10% improvements each
>year.
It never did 100% per year sustained; what has changed is that instead of large jumps every year (or even few years) we now see many small increases. For instance, the original Pentium was launched in early 1993 at 60/66 MHz. It took more than a year for a new one to hit the market at 90 Mhz, then another year to hit 120 Mhz and for 166 MHz you had to wait until a full three years (till 1996) from the original 66 MHz one. Now we see 5-10% improvements every so many months.
>What happened to the "we can go to 8Ghz with the 90nm
>process..." yada yada yada.
It referred to netburst (but I doubt the process node was mentioned), and netburst hit a thermal brick wall.
>And the motherboards are just as guilty of the lack of pure
> performance improvemen
Motherboard makers just implement, they never make the leaps, they cant even if they could. It takes chipsets and cpu's made by Intel/AMD/nVidia/via,.. but yes, FSB dont scale as fast cpus, and neither does the RAM. That is nothing new though; in the 486 (pre DX2) era CPU, FSB and RAM all ran synchronous. Ever since the cpu has outpaced the other 2 and that wont change ever. You could either blame FSB/RAM tech, or credit cpu improvements, but its not without reason AMD got rid of the FSB bottleneck, and likewise I expect embedded RAM to become the next step. in fact, large L2/3 caches could be considered just that already.
>The only tech industry that shows signs of improvement
>comes from Graphics processors (nVidia and ATI).
really ? All they do is make their GPUs wider with each respin, its very much like making multicore CPUs, except that 3D apps can actually make good use of them. Furthermore its about time nVidia and ATI start taking power consumption seriously. I read elsewhere that current high end videocards consume up to 50W *iddle*, there is no excuse for that.
>Is the mass migration of jobs to bio-tech a true indication
> of the state of the tech industry?
LOL, no. It is true however that the market is being commodotized more and more. just like we dont care much anymore about the "performance" of our microwave ovens, the number of ppl that need/want ever faster performance is shrinking rapidly. But we are certainly not anywhere near the end of technology scaling yet.
>Have AMD/Intel out sourced so much so that innovation is no
> longer possible?
They outsource next to nothing, and this has absolutely nothing to do with anything.
= The views stated herein are my personal views, and not necessarily the views of my wife. =
When it comes to the corporate world then PC does not need to be faster for the average end user.
Most companies are deploying more and more server based applications. Desktop administration is very costly.
For each PC we have to buy (we have 35000 desktops in our organization), the admistration costs go up with each application running. (In some cases the admin costs per year are higher than the hardware costs)
Where we do have high performance needs, the applications are multitheaded, hence we use dual CPU/dual core or in some cases even 4 socket systems.
The above approach is even starting to filter though to the home desktop, with more and more applications being served through set top boxes,i.e. email and web surfing via your TV, games via your Xbox/PS2, etc. There will no doubt come a time where basic Office type applications will be network based.
I personally welcome this as I won't have to spend hours getting rid of all the crap that is downloaded to my PC, constantly reinstalling when things to wrong (drivers,software, etc, etc)
BTW anybody else see how a browser is becoming more and more like an X-client, if only windows had gone that way?)
However in the mean time we have millions of home users complaining that their PCs are running slow. They generally don't have the time,knowledge or inclination to delete all of the rubbish mounting up on their desktop (adware, etc, etc). It is far easier for them to buy a new PC. A dual core will benefit them far more than a high frequency single core chip as they won't notice the crap until both core become saturated.
Most companies are deploying more and more server based applications. Desktop administration is very costly.
For each PC we have to buy (we have 35000 desktops in our organization), the admistration costs go up with each application running. (In some cases the admin costs per year are higher than the hardware costs)
Where we do have high performance needs, the applications are multitheaded, hence we use dual CPU/dual core or in some cases even 4 socket systems.
The above approach is even starting to filter though to the home desktop, with more and more applications being served through set top boxes,i.e. email and web surfing via your TV, games via your Xbox/PS2, etc. There will no doubt come a time where basic Office type applications will be network based.
I personally welcome this as I won't have to spend hours getting rid of all the crap that is downloaded to my PC, constantly reinstalling when things to wrong (drivers,software, etc, etc)
BTW anybody else see how a browser is becoming more and more like an X-client, if only windows had gone that way?)
However in the mean time we have millions of home users complaining that their PCs are running slow. They generally don't have the time,knowledge or inclination to delete all of the rubbish mounting up on their desktop (adware, etc, etc). It is far easier for them to buy a new PC. A dual core will benefit them far more than a high frequency single core chip as they won't notice the crap until both core become saturated.
>BTW anybody else see how a browser is becoming more and more
>like an X-client,
Ahem.. no. Last time I heard someone say something like that, it was Bill Gates who feared Netscape might become the next OS
. Ever since I thought a browser was just... a browser. Of course meanwhile Oracle, Sun and others have tried to reinvent the thin client, and we all know what stellar success they have had with them... So no, I dont believe in running thin clients with networked/browser based apps as a general solution (it works wonders for netbanking or email though). If that made sense, many users would be behind an cheapo ARM or whatever based thin client, yet no one is.. i think ?
>It is far easier for them to buy a new PC
So true..
>A dual core will benefit them far more than a high
>frequency single core chip as they won't notice the crap
>until both core become saturated
LOL.. DC as a solution to spyware..! Im not even sure it would help, most PC slowdowns Ive seen are a result of registry bloat and file fragmentation. Im not sure if a second core would be any good for either problem, but I doubt it. A ramdrive OTOH....
= The views stated herein are my personal views, and not necessarily the views of my wife. =
>like an X-client,
Ahem.. no. Last time I heard someone say something like that, it was Bill Gates who feared Netscape might become the next OS
. Ever since I thought a browser was just... a browser. Of course meanwhile Oracle, Sun and others have tried to reinvent the thin client, and we all know what stellar success they have had with them... So no, I dont believe in running thin clients with networked/browser based apps as a general solution (it works wonders for netbanking or email though). If that made sense, many users would be behind an cheapo ARM or whatever based thin client, yet no one is.. i think ?>It is far easier for them to buy a new PC
So true..
>A dual core will benefit them far more than a high
>frequency single core chip as they won't notice the crap
>until both core become saturated
LOL.. DC as a solution to spyware..! Im not even sure it would help, most PC slowdowns Ive seen are a result of registry bloat and file fragmentation. Im not sure if a second core would be any good for either problem, but I doubt it. A ramdrive OTOH....
= The views stated herein are my personal views, and not necessarily the views of my wife. =
Is there anyone at Tom's Hardware that actually has a clue what they are talking about?
This place is really a waste of space -- more on the lines with The Star or Inquirer.
Last response from me -- best of luck to you all, you'll need it.
This place is really a waste of space -- more on the lines with The Star or Inquirer.
Last response from me -- best of luck to you all, you'll need it.
Unolocogringo
Illustrious
- Dec 31, 2007
- 8,022
- 810
- 38,390
Damn it sounds like you swallowed hook, line, sinker,rod& reel,boat and trolling motor.
I had no Idea Intels PR Marketing department were that good.
I have this bridge you might be interested in .........
I aint signing nothing!!!
I had no Idea Intels PR Marketing department were that good.
I have this bridge you might be interested in .........
I aint signing nothing!!!
slvr_phoenix
Splendid
- Dec 31, 2007
- 6,157
- 1
- 25,780
Oh, I don't know. I think we'd enjoy that waste of time, since you seem to get everything else so humorously wrong.
It does if you're smart enough to assign priorities to them.
**ROFL** Wow. When you're wrong, you go all out. First, as everyone knows, not all of those services are active. Only about half of mine are even started. Most are set to manual and haven't been started yet. So you're already significantly wrong.
Second, if you really want an accurate thread count, give your PC the good 'ol 3 finger salute (CTRL+ALT+DEL) and go to Task Manager. (Or use the WinXP shortcut of CTRL+SHIFT+ESC to get there faster.) The Performance tab lists the total thread count. I've got 465 running right now. :O Or you can use the Processes tab to view the CPU Time and Thread count to get an acurate idea of what is <i>really</i> going on.
The point is not to have one CPU per thread. That'd be daft, especially as most hardly use any CPU resources. What the point is, is to have one CPU running the intensive task, and to have the other CPU running everything else. It's cheap, easy, and effective.
This is a funny question, as IMHO, the mass market of a dualie box at an affordable price has only been a very short while now, and even <i>that</i> is still rather debatable.
یί∫υєг ρђœŋίχ <i>The <font color=red><b>Devil</b></font color=red> is in the details.</i>
@ 198K of 200K!
slvr_phoenix
Splendid
- Dec 31, 2007
- 6,157
- 1
- 25,780
Everyone but you by all appearances...
I'll believe it when I see it. After all, nearly everything else that you've said has been wrong.
یί∫υєг ρђœŋίχ <i>The <font color=red><b>Devil</b></font color=red> is in the details.</i>
@ 198K of 200K!
This thread is hilarious. I don't know why I hadn't read it until now.. :lol:
I'm glad we're not hiring software engineers of his caliber. I think a lot of Daily WTFery will flow from his fingers when he graduates (since I think I read in amongst all that that he thinks he'll become a programmer)
---
<pre> (\_/)
|~~~~~|======
|_____| This was bunny. He was tasty.
/\/\/\/\</pre><p>
I'm glad we're not hiring software engineers of his caliber. I think a lot of Daily WTFery will flow from his fingers when he graduates (since I think I read in amongst all that that he thinks he'll become a programmer)
---
<pre> (\_/)
|~~~~~|======
|_____| This was bunny. He was tasty.
/\/\/\/\</pre><p>
xxsk8er101xx
Splendid
- Dec 31, 2007
- 3,967
- 4
- 22,815
That's cheap though ... i think the law was meant for single processors not slopping on a second core.
I believe moores law had nothing to do with # of transistors but infact actual performance. You can make a cpu with 1 billion transistors and have it perform like a pentium 200.
<A HREF="http://www.anandtech.com/mysystemrig.html?id=9933" target="_new"> My Rig </A>
I believe moores law had nothing to do with # of transistors but infact actual performance. You can make a cpu with 1 billion transistors and have it perform like a pentium 200.
<A HREF="http://www.anandtech.com/mysystemrig.html?id=9933" target="_new"> My Rig </A>
xxsk8er101xx
Splendid
- Dec 31, 2007
- 3,967
- 4
- 22,815
this is just a temporary slow down. It happend before when the pentium chip came out. There was a long period where nothing new came up and the top chip was the pentium 200. Intel then came out with the pentium pro. Then later on the pentium 2 came out. It wasn't until the later pentium 3's and pentium 4's where the performance of CPU's exploded.
Well its catching up to us now. There was a huge explosion in speed and as a result there is a slow down. It's simply normal because there tends to be technological hurtles. I wouldn't be surprised if you saw a new socket type next year for both AMD and Intel.
<A HREF="http://www.anandtech.com/mysystemrig.html?id=9933" target="_new"> My Rig </A>
Well its catching up to us now. There was a huge explosion in speed and as a result there is a slow down. It's simply normal because there tends to be technological hurtles. I wouldn't be surprised if you saw a new socket type next year for both AMD and Intel.
<A HREF="http://www.anandtech.com/mysystemrig.html?id=9933" target="_new"> My Rig </A>
It's <i>never</i> been about performance. I always thought it was more about the number of transistors, but according to <A HREF="http://www.xbitlabs.com/articles/editorial/display/moore.html" target="_new">X-bit labs</A>, it's more about the economics of building larger and larger ICs and how that would make things more affordable than they were at the time.
Just did a little googling.
A vaguely interesting read.---
<pre> (\_/)
|~~~~~|======
|_____| This was bunny. He was tasty.
/\/\/\/\</pre><p>
slvr_phoenix
Splendid
- Dec 31, 2007
- 6,157
- 1
- 25,780
<pre><font color=orange> ∩_∩
Ω Ω
(=¥=)</font color=orange> - Cedrik wonders if maybe the ah heck said something that was true after all...<font color=orange>
_Ū˘Ū_</font color=orange></pre><p>
یί∫υєг ρђœŋίχ <i>All your <b><font color=red>Devil</font color=red></b> are belong to us.</i> - Cedrik
@ <b>199</b>K of 200K!
There is no set law... There is only a belief or an understanding, all of everyones points are valid.
<font color=red>"Battling Gimps and Dimbulbs HERE at THGC"</font color=red>
"<font color=blue> Wusy</font color=blue> <-Professional sheep banger"
<font color=red>"Battling Gimps and Dimbulbs HERE at THGC"</font color=red>
"<font color=blue> Wusy</font color=blue> <-Professional sheep banger"
TRENDING THREADS
-
-
-
News Introducing the Tom’s Hardware Premium Beta: Exclusive content for members
- Started by Admin
- Replies: 30
-
Question Futureproofing for next 3-5 years: What should I upgrade to to be able to run BF6, GTA 6 etc?
- Started by mechagran
- Replies: 5
-
-
Discussion What's your favourite video game you've been playing?- Started by amdfangirl
- Replies: 4K
Facebook
Twitter
Instagram