Last message on previous page: Archived from groups: alt.comp.hardware.overclocking.amd,comp.sys.ibm.pc.hardware.chips,uk.comp.homebuilt,alt.comp.hardware (More info?)
On Fri, 10 Sep 2004 15:12:39 +0200, Kai Harrekilde-Petersen wrote:
> Daniel James <wastebasket@nospam.aaisp.org> writes:
> >> It's an interesting question ... though not as interesting as it was in >> 1983. I used to have a rule-of-thumb for estimating the "bittedness" of >> CPUs that went something like this:
>>
>> The 8088 had an 8-bit external data bus, 16-bit general-purpose registers, >> 16-bit stack pointer, and 20-bit addressing range (being a bit generous, >> considering the segmented addressing model). Taking an average of those >> four gives 15 bits. Near enough.
>>
>> The 8086 was the same apart from the 16-bit data bus (and, yes, the longer >> instruction prefetch queue - I'd forgotten about that) so let's call it 17 >> bits.
>>
>> Compare that with the Z80 (8+16[1]+16+16=14 bits) the 6502 beloved of >> Apple II and PET user (8+8+8+16=10 bits) and the 68000 (32+32+32+24=30 >> bits). Considering that these chips were all available when the PC was >> designed: I know which chip *I'd* have used!
> > It is well publized that IBM chose a lower-end CPU for the PC in order
> not to compete with other business divisions.
Not really. The reason the 8088 was chosen was for cost. A 16b bus would
have doubled the cost of the bus and the minimum memory configuration. Remember, memory was *expensive*. ...as was SSI/MSI TTL and packaging.
Archived from groups: comp.sys.ibm.pc.hardware.chips (More info?)
keith wrote:
> Generally the "bitness" of a processor is it's linear (indexed)
> address range or the width of its FX register file (which are usually
> the same thing).
What is the FX register file? Is the width of a register file defined
as the width of every register in the RF?
Archived from groups: comp.sys.ibm.pc.hardware.chips (More info?)
On Fri, 10 Sep 2004 17:23:05 +0200, Grumble wrote:
> keith wrote:
> >> Generally the "bitness" of a processor is it's linear (indexed)
>> address range or the width of its FX register file (which are usually
>> the same thing).
> > What is the FX register file?
FiXed point registers == General purpose registers.
> Is the width of a register file defined as the width of every register
> in the RF?
I've never seen a processor with rifferent length registers in the file,
but I suppose it's possible. I have no idea why one would want to do such
a thing.
Archived from groups: alt.comp.hardware.overclocking.amd,comp.sys.ibm.pc.hardware.chips,uk.comp.homebuilt,alt.comp.hardware (More info?)
On Fri, 10 Sep 2004 12:58:21 +0100, Daniel James wrote:
> Compare that with the Z80 (8+16[1]+16+16=14 bits) the 6502 beloved of > Apple II and PET user (8+8+8+16=10 bits) and the 68000 (32+32+32+24=30 > bits). Considering that these chips were all available when the PC was > designed: I know which chip *I'd* have used!
> Me too, but the 68000 only had a 16 bit data bus. And at the time, cpu's
were rated rated by the data bus width. Then 68000 which was defined by
Motorola as a 16 bit cpu is now define as a 32bit cpu.:-)
8088 (8/16) was defined by Intel as an 8 bit cpu. Today it's defined as a
16bit CPU. One can no longer take anything for granted as the companys
have started fudging and cludging numbers for marketing hype. Here's the
best BS I've ever seen. NEC compares their NEAX 2000 pbx to Toshiba's
(forgot the name) calling their processor a powerful 32bit bit compared to
Toshibas 16bit. NEC was using a 486SLC (8 bit data bus) while Toshiba used
a 68000 (16 bit data bues).:-)
Now if you have ever used the 2, you know what I mean.
Archived from groups: alt.comp.hardware.overclocking.amd,comp.sys.ibm.pc.hardware.chips,uk.comp.homebuilt,alt.comp.hardware (More info?)
On Fri, 10 Sep 2004 11:06:51 -0400, keith wrote:
> On Fri, 10 Sep 2004 15:12:39 +0200, Kai Harrekilde-Petersen wrote:
> >> It is well publized that IBM chose a lower-end CPU for the PC in order
>> not to compete with other business divisions.
> > Not really. The reason the 8088 was chosen was for cost. A 16b bus would
> have doubled the cost of the bus and the minimum memory configuration. > Remember, memory was *expensive*. ...as was SSI/MSI TTL and packaging.
Hmmm... Cost played a part in ti, but the 80xx architecture was chosen
because that's the only micro architecture the developers were familiar
with and they wanted to get it out fast. It was just a matter of months
before someone came out with a real 16 bit cpu, the 8086, and it wasn't
IBM.
Archived from groups: alt.comp.hardware.overclocking.amd,comp.sys.ibm.pc.hardware.chips,uk.comp.homebuilt,alt.comp.hardware (More info?)
In comp.sys.ibm.pc.hardware.chips Franklin <franklin_lo@mail.com> wrote:
> A Barton 2500+ (with maybe an Asus A78NX mobo) is more than enough > power for me but am I buying into obsolescence? Athlon64 is where > the growth will be and furture residual values will be higher than > for Barton.
Future residual values? These are computers, not real estate -- odds are by
the time you need to get a new machine, you'll have trouble getting more
than beer money for the machine.
I've had trouble giving away some of the machines I've had.
Archived from groups: comp.sys.ibm.pc.hardware.chips,uk.comp.homebuilt,alt.comp.hardware (More info?)
borolad@myowseintheboro.org wrote:
> *'``'*:-.,_,.-:*'``' ¸ô¶ó - Cull the O/T Shite '``'*:-.,_,.-:*'``'*
This makes over ten piles you've dropped already!
Do you think your *Shite* doesen't stink? ;-(
VWW
-- A foolish consistency is the
hobgoblin of little minds,........
Ralph Waldo Emerson
(Microsoft programmer's manual.)
Archived from groups: alt.comp.hardware.overclocking.amd,comp.sys.ibm.pc.hardware.chips,uk.comp.homebuilt,alt.comp.hardware (More info?)
On Fri, 10 Sep 2004 18:00:03 +0000, Wes Newell wrote:
> On Fri, 10 Sep 2004 11:06:51 -0400, keith wrote:
> >> On Fri, 10 Sep 2004 15:12:39 +0200, Kai Harrekilde-Petersen wrote:
>> >>> It is well publized that IBM chose a lower-end CPU for the PC in order
>>> not to compete with other business divisions.
>> >> Not really. The reason the 8088 was chosen was for cost. A 16b bus would
>> have doubled the cost of the bus and the minimum memory configuration. >> Remember, memory was *expensive*. ...as was SSI/MSI TTL and packaging.
> > Hmmm... Cost played a part in ti, but the 80xx architecture was chosen
> because that's the only micro architecture the developers were familiar
> with and they wanted to get it out fast. It was just a matter of months
> before someone came out with a real 16 bit cpu, the 8086, and it wasn't
> IBM.
First of all, the question was "why the 8088". The IBM originally was
going to go with the 8085 (which is not in any way the same architecture
as the 8086/8), which an entirely different group in IBM used at about the
same time for the "DataMaster". IBM had no real experience with either,
since up until then they generally used their own microprocessors. As the
story goes, Bill Gates convinced IBM ESD that 16bits was the way to go.
Once that decision was made, Intel was chosen primarily because it
"wasn't Motorola" and the 8088 was chosen because an 8088 system was much
cheaper than one using the 8086.
Second, the 8088 is by any reasonable definition a "real" 16-bit
processor. It has *exactly* the same architecture as the 8086, so if you
consider the 8086 a "real" 16b processor, then you must also admit that
the 8088 is.
It was more than a few months before the 8086 clones started.
Q had their clone out by summer though.
Archived from groups: alt.comp.hardware.overclocking.amd,comp.sys.ibm.pc.hardware.chips,uk.comp.homebuilt,alt.comp.hardware (More info?)
On Fri, 10 Sep 2004 17:53:53 +0000, Wes Newell wrote:
> On Fri, 10 Sep 2004 12:58:21 +0100, Daniel James wrote:
> >> Compare that with the Z80 (8+16[1]+16+16=14 bits) the 6502 beloved of >> Apple II and PET user (8+8+8+16=10 bits) and the 68000 (32+32+32+24=30 >> bits). Considering that these chips were all available when the PC was >> designed: I know which chip *I'd* have used!
>> > Me too, but the 68000 only had a 16 bit data bus. And at the time, cpu's
> were rated rated by the data bus width.
Nonsense. At the time bitness was a measure of the accumulator width. The data bus had no bearing on it. Bus width is an implementation
detail. DO you consider the Pentium a 64bit processor?
> Then 68000 which was defined by
> Motorola as a 16 bit cpu is now define as a 32bit cpu.:-) 8088 (8/16)
> was defined by Intel as an 8 bit cpu.
No it was not. The 8088 was always defined as a 16 bit processor. It was
marketed as a cost-reduced (at the system level) 8086.
> Today it's defined as a 16bit CPU.
> One can no longer take anything for granted as the companys have started
> fudging and cludging numbers for marketing hype. Here's the best BS I've
> ever seen. NEC compares their NEAX 2000 pbx to Toshiba's (forgot the
> name) calling their processor a powerful 32bit bit compared to Toshibas
> 16bit. NEC was using a 486SLC (8 bit data bus) while Toshiba used a
> 68000 (16 bit data bues).:-)
> Now if you have ever used the 2, you know what I mean.
Archived from groups: alt.comp.hardware.overclocking.amd,comp.sys.ibm.pc.hardware.chips,uk.comp.homebuilt,alt.comp.hardware (More info?)
NASCAR from EAsports is asome with 3 montiors "Dave" <dave@hotmail.com> wrote in message news:hFh0d.25850$D7.12351@news-server.bigpond.net.au...
> Snip 8<
>
>> I can't imagine AGP entirely disappearing within a year or two. There >> will still be a big market for AGP upgrade cards in the near future. >> Heck, you can still buy a GeForce 5200 or Radeon 9200 in PCI.
>
> Yes, very handy for adding extra monitors to the computer. Too bad very > few games make use of extra monitors.
>
Archived from groups: alt.comp.hardware.overclocking.amd,comp.sys.ibm.pc.hardware.chips,uk.comp.homebuilt,alt.comp.hardware (More info?)
On Fri, 10 Sep 2004 15:29:44 -0400, keith wrote:
> Second, the 8088 is by any reasonable definition a "real" 16-bit
> processor. It has *exactly* the same architecture as the 8086, so if you
> consider the 8086 a "real" 16b processor, then you must also admit that
> the 8088 is. > No one cares how you define, just as no one cares how I define, but the
fact is that Intel itself defined it as an 8 bit CPU when it first came
out. I had this same arguement a couple of years ago. Someone with a
scanner finally proved me right. I'm still willing to bet though.:-)
> It was more than a few months before the 8086 clones started. Q had
> their clone out by summer though.
I consider 3-6 to be a few. Now I'm not certain, but I'm pretty sure 8086
clones showed up within that time period. And if it were 8 months, so what.
Archived from groups: alt.comp.hardware.overclocking.amd,comp.sys.ibm.pc.hardware.chips,uk.comp.homebuilt,alt.comp.hardware (More info?)
Dave wrote:
> "gaffo" <gaffo@usenet.net> wrote in message > news:aQ60d.9557$yp2.8676@newssvr30.news.prodigy.com...
> >>Wes Newell wrote:
>>
>>
>>>On Thu, 09 Sep 2004 03:03:11 +0000, gaffo wrote:
>>>
>>>
>>>
>>>>Feel free to *plonk* me it my sign offends.
>>>>
>>>
>>>Doesn't offend me.
>>
>>
>>
>>thats good.
>>
>>
>>
>>
>>>It just shows your stupidity and lack of consideration
>>>for others.
>>
>>
>>
>>
>>My civic duty to my Nation as a citizen is more important then my >>civility - Liberty is on the line.
> > > How about the rest of the world? Do you actually think everybody else on > the whole planet actually cares about your particular political belief?
> > Why don't you target Bush's electorate with your views instead of the rest > of the world (those of us that have NG/internet access, anyway)?
> > Dave > >
good point Dave..............sadly I still must follow what I feel is my duty, even if it offends non-US folks.
Archived from groups: alt.comp.hardware.overclocking.amd,comp.sys.ibm.pc.hardware.chips,uk.comp.homebuilt,alt.comp.hardware,alt.comp.periphs.mainboard.tyan (More info?)
> >> it'll be interesting so see what we can go to when the 9 micron
Opterons
> >> show up in the near future
>
> > You mean .09um (90nm)?
>
> yes, 90nm. isn't that the same as 9 micron?
Archived from groups: alt.comp.hardware.overclocking.amd,comp.sys.ibm.pc.hardware.chips,uk.comp.homebuilt,alt.comp.hardware (More info?)
On Fri, 10 Sep 2004 22:36:10 +0000, Wes Newell wrote:
> On Fri, 10 Sep 2004 15:29:44 -0400, keith wrote:
> >> Second, the 8088 is by any reasonable definition a "real" 16-bit
>> processor. It has *exactly* the same architecture as the 8086, so if you
>> consider the 8086 a "real" 16b processor, then you must also admit that
>> the 8088 is. >> > No one cares how you define, just as no one cares how I define, but the
> fact is that Intel itself defined it as an 8 bit CPU when it first came
> out. I had this same arguement a couple of years ago. Someone with a
> scanner finally proved me right. I'm still willing to bet though.:-)
Ind I say you're wrong! Intel defined the 8088 as a 16b processor. It
was sold as a cheapie, because it had an 8b bus.
>> It was more than a few months before the 8086 clones started. Q had
>> their clone out by summer though.
> > I consider 3-6 to be a few. Now I'm not certain, but I'm pretty sure
> 8086 clones showed up within that time period. And if it were 8 months,
> so what.
Try 10-12. I'm not sure the original Q's had an 80886 either. ...don't
remember. The performance difference wasn't staggering in any case.
