Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

processor

Status
Not open for further replies.
Overclocked said:
Its SSOI, but earlier procs used SOI. **broken link removed**

Speaking of Tech, Did anyone catch that AOL released personal info on searches performed by users?
http://www.theregister.co.uk/2006/08/07/aol_search_logs/
It doesn't matter. The 'problems' you speak of do not show themselves in any way shape or form in Intel's silicon. For you to call Intel Silicon 'low quality' just because it doesn't have some feature, which has no impact whatsoever on the actual product, is simply idiotic.

So, if Intel can beat AMD in EVERY SINGLE POSSIBLE WAY using their 'crappy' silicon process, then what does that really say about AMDs products? Even with all the svelte and sexy technology that they supposedly use, they can't turn out a product that competes in the same arena.
 
Marks256 said:
No, it is not. Here is the link;
https://news.com.com/Chip+breaks+speed+record+in+deep+freeze/2100-1006_3-6085568.html?tag=nefd.top

It was possable by bringing the CPU just above absolute zero. The CPU probably isn't a fully functional CPU, but it is a start. It was made using germainium mixed with the silicon.

Yes, but there are some questions to be asked ..
Is operating your cpu in such a low temperature plausible for a desktop pc ? i think not ...

Second, The chip itself runs at those awesome speeds. The chip is tiny and electricity can get where it needs to be in that tiny amount of time ...

However , if it were a full computer, it would need data to work on ... from memory. Preferably fast enough so the cpu doest not have to wait for data to arrive ...

How are you going to connect RAM and IO to that 500 Ghz cpu ?
if you put those parts 1cm away from the cpu , electricty can't reach it in time ..

They are already hitting a wall tbh, why do you think all major cpu makers are moving toward parallellism ? Because they can't get the speeds up any higher.
 
poopeater said:
It doesn't matter. The 'problems' you speak of do not show themselves in any way shape or form in Intel's silicon. For you to call Intel Silicon 'low quality' just because it doesn't have some feature, which has no impact whatsoever on the actual product, is simply idiotic.

So, if Intel can beat AMD in EVERY SINGLE POSSIBLE WAY using their 'crappy' silicon process, then what does that really say about AMDs products? Even with all the svelte and sexy technology that they supposedly use, they can't turn out a product that competes in the same arena.


It realy depends on the market.
At this moment, in the desktop and laptop market, using both single and dual core tech intel is no doubt the leader.
Even the lowest end core2 cpu's can rival amd's FX62 top desktop part.
SOI is a nice tech allowing AMD to push their cpu's higher, but core2 definately proves there are other ways. Core overclocks way higher then any AMD cpu to date.

As for AMD's integrated memory controller. It is the way of the future, intel already acknowledged that and they are working on their own IMC, but in the mean time FSB tech , FOR DESKTOPS, isn't dead yet. Again core 2 proves this.

However, in the server market, opeteron is and will remain king.
Simply because current intel tech isn't scalable. Kentsfield, intel's quad core server chip, returns to the old and slow "glue 2 conroe cores on the FSB" way. a 2way 8 core system will have 2 sockets for 2 kentsfield cpu's wich are again merged on the front side bus ..
Those awesome kentsfield cores will be starved to death on an overcrowded FSB.

AMD has ccHT to connect cpu's together trough a ringbus, wich makes the systems scale very good.
Socket F K8L is even rumoured to have ccHT2, allowing p2p connection of upto 16cpu's (quad core ? = 64cores )
there's no way a FSB based system can match this.
 
Last edited:
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top