SI
SI
discoversearch

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor.  We ask that you disable ad blocking while on Silicon Investor in the best interests of our community.  If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.
Technology Stocks : Intel Corporation (INTC) -- Ignore unavailable to you. Want to Upgrade?


To: Paul Engel who wrote (46215)1/22/1998 12:26:00 AM
From: FJB  Respond to of 186894
 
Excellent post Paul. Thanks. Bob (EOM)



To: Paul Engel who wrote (46215)1/22/1998 12:53:00 AM
From: Elmer  Read Replies (1) | Respond to of 186894
 
Paul, the PCI 2.1 specification allows for 66mhz , 64 bit data paths. As you mentioned, only DEC is doing this AFAIK. Intel has mentioned that AGP is a superset of this spec. It will perform these 2.1 operations, but only with a 32 bit bus. In addition, it provides for 66mhz "double pumped" operation which allows for 132mhz transfers.
The net effect of (32 bits x 132mhz) = (64bits x 66mhz) but with lower pincount. As for AGP, as it is currently implemented, AGP is actually a port, even though it looks like a bus. It allows for only 1 device. There is no allowance for multiple devices to arbitrate for the port. I don't know if AGP2 will change to a bus. Could the current AGP port be used in a server for non-graphics applications? Hey, that's a good idea! I wonder if Intel is looking into this?

Cheers,

EP



To: Paul Engel who wrote (46215)1/22/1998 1:11:00 AM
From: Jules V  Respond to of 186894
 
Paul,
Thanks for the comprehensive reply.

I just realized I could do with a basic "Bus" refresher.
Eg. basic diff between the AGP port, BX 100, and PCI again.
Is AGP at all similar concept to "local bus".

Thanks again.



To: Paul Engel who wrote (46215)1/22/1998 2:57:00 AM
From: Barry Grossman  Read Replies (2) | Respond to of 186894
 
Paul,

Excellent info.

Where do you get this stuff?

Barry



To: Paul Engel who wrote (46215)1/22/1998 8:54:00 AM
From: Barry A. Watzman  Read Replies (1) | Respond to of 186894
 
Will it be possible to run current PII chips on the BX chipset with a 100 MHz bus ? For example, a 300MHz PII currently uses a 66 MHz clock with a 4.5 clock-to-CPU speed multiplier. Does this chip have a 3x multiplier option, and if so, could you run it with a 100 MHz clock and a 3x multiplier ? Another case, run a PII 233 (normally 66 MHz with a 3.5 multiplier) at 250 MHz (a slight overclock) using a 100 MHz clock and a 2.5 multiplier. Will any of these be officially supported ? Separate question, ignoring official specifications, will any of these combinations be physically possible ?



To: Paul Engel who wrote (46215)1/22/1998 11:17:00 AM
From: Kirk ©  Read Replies (2) | Respond to of 186894
 
We've specualted on the "Killer App" to drive people to upgrade. I think I have it. RSI prevention vis a vis voice recognition. We're just experimenting with it and it seems to take 64M on W95 and 82M on NT, a great sound card and a "reasonable" Pentium to just get running. My guess, as the software gets better, more processor speed will allow it to run better as with most applications.

Besides being long Intel, I wonder if there are any other investments that can take advantage of this? MU seems obvious.

regards
Kirk out

"hello Computer" - Commander Scott talking into a mouse after being handed one and told to use it to interface with a PC.

More FREE Intel discussion at suite101.com



To: Paul Engel who wrote (46215)1/22/1998 1:15:00 PM
From: StockMan  Respond to of 186894
 
Paul,
Re -- The ability of NT to handle a contiunual burst of data, coming in on an adapter card at 1 GigaBit/Sec, is the first bottleneck.

This is true, however the I2O (batch interrupt processing) and NT 5.0 could alleviate this.

Stockman