clipka <ano### [at] anonymousorg> wrote:
>
> The image was most likely rendered on a 386 DX-40 machine (without
> mathematical co-processor), using /some/ DOS version of POV-Ray
> (presumably 2.0). As for the rendering time, I just recall it to have
> been /awfully/ long.

I thought the DX machines were the ones with the co-processor, my first PC was a
386 SX-33 and I remember complaining to the person who built it as I had asked
for a DX with a co-processor and 4MB of memory. I remember him saying nothing
has been written to use the co-processors so he did not think there was any
point in adding one. Also asking me why I needed 4MB of memory, he obviously was
not aware of POV-Ray.

Unfortunately I do not have my first image any longer so cannot post it. I have
the files on an old 3.25 floppy disk but do not have anything with a floppy disk
drive to get hold of it..

Sean

Post a reply to this message

On 8/3/2011 10:40, s.day wrote:
> do not have anything with a floppy disk
> drive to get hold of it..

USB FDD FTW.  Just in case you ever *really* needed to get to it and didn't 
know they made such things.

-- 
Darren New, San Diego CA, USA (PST)
   How come I never get only one kudo?

Post a reply to this message

On Wed, 03 Aug 2011 13:40:02 -0400, s.day wrote:

> I thought the DX machines were the ones with the co-processor, my first
> PC was a 386 SX-33 and I remember complaining to the person who built it
> as I had asked for a DX with a co-processor and 4MB of memory. I
> remember him saying nothing has been written to use the co-processors so
> he did not think there was any point in adding one. Also asking me why I
> needed 4MB of memory, he obviously was not aware of POV-Ray.

Nope, the DX had a 32-bit data bus, the SX only had a 16-bit data bus 
(but a 32-bit instruction set).  The 24 bit address pins meant the SX 
could only handle 16 MB of RAM total, as opposed to 4 GB on the DX models 
(of course, no 386 ever had that much memory in it; my DX was maxed out 
at 16 MB).

And yes, obviously he wasn't aware of POV-Ray; my first serious raytracing 
system was a Northgate 80386DX-33 with a 80387 math coprocessor (I was an 
aeronautical engineering student at the time, so paid for the mathco as 
an add-on option).  IIRC, POV-Ray at the time would take advantage of a 
math coprocessor if it was present.

Jim

Post a reply to this message

s.day wrote:

> I thought the DX machines were the ones with the co-processor

that was true for the 486

> Unfortunately I do not have my first image any longer so cannot post it. I have
> the files on an old 3.25 floppy disk but do not have anything with a floppy disk
> drive to get hold of it..

I put two old FDDs (5.25 + 3.5) in my Windows 7 64-bit
hexacore only to find out BIOS doesn't support more than
one FDD. But it sure looks retro ;)

Post a reply to this message

Jim Henderson <nos### [at] nospamcom> wrote:
> Nope, the DX had a 32-bit data bus, the SX only had a 16-bit data bus
> (but a 32-bit instruction set).  The 24 bit address pins meant the SX
> could only handle 16 MB of RAM total, as opposed to 4 GB on the DX models
> (of course, no 386 ever had that much memory in it; my DX was maxed out
> at 16 MB).
>
> And yes, obviously he wasn't aware of POV-Ray; my first serious raytracing
> system was a Northgate 80386DX-33 with a 80387 math coprocessor (I was an
> aeronautical engineering student at the time, so paid for the mathco as
> an add-on option).  IIRC, POV-Ray at the time would take advantage of a
> math coprocessor if it was present.
>
> Jim

Wish my memory was as good as yours. On a more positive note, I actually found
the files in an old zip file stashed away on my work PC. Funny to re-trace this
image now, I think it took about 2 weeks to draw on my 386-SX 33Mhz PC at
320x240 and even with AA0.05 now is only 1m49s on my current PC. I did not even
need to make any code changes (a couple of version comments had to be removed as
this was for version 2.0).

Glad to have revived it without needing to buy a USB floppy drive.

I am sure this was not my very first image but it was the first proper one I
created once I had learnt how to use POV.

Sean

Post a reply to this message

Attachments:
Download 'skel.jpg' (104 KB)

Preview of image 'skel.jpg'

On Wed, 03 Aug 2011 19:09:29 -0400, s.day wrote:

> Wish my memory was as good as yours. 

I didn't remember the technical differences, I had to verify them 
(Wikipedia proved a good resource for this info), but did remember that 
the 80386 didn't include an on-board coprocessor in any incarnation.  
That was, as Christian stated, a differentiator for the 80486 processor 
line.

> On a more positive note, I actually
> found the files in an old zip file stashed away on my work PC. Funny to
> re-trace this image now, I think it took about 2 weeks to draw on my
> 386-SX 33Mhz PC at 320x240 and even with AA0.05 now is only 1m49s on my
> current PC. I did not even need to make any code changes (a couple of
> version comments had to be removed as this was for version 2.0).

That's pretty cool, I wish I could find my first published render (the 
first was something like a standard RSOCP to see what could be done), it 
was a cribbage board IIRC.  I know Chris Cason still has the listing from 
the old CompuServe POVRAY forum.  The cards were scans of a deck I 
purchased in Russia when I was there in 1988.

Jim

Post a reply to this message

"s.day" <s.d### [at] uelacuk> wrote:
> clipka <ano### [at] anonymousorg> wrote:
> >
> > The image was most likely rendered on a 386 DX-40 machine (without
> > mathematical co-processor), using /some/ DOS version of POV-Ray
> > (presumably 2.0). As for the rendering time, I just recall it to have
> > been /awfully/ long.
>
> I thought the DX machines were the ones with the co-processor, my first PC was a
> 386 SX-33 and I remember complaining to the person who built it as I had asked
> for a DX with a co-processor and 4MB of memory. I remember him saying nothing
> has been written to use the co-processors so he did not think there was any
> point in adding one. Also asking me why I needed 4MB of memory, he obviously was
> not aware of POV-Ray.
>
> Unfortunately I do not have my first image any longer so cannot post it. I have
> the files on an old 3.25 floppy disk but do not have anything with a floppy disk
> drive to get hold of it..
>
> Sean

Actually the 386 DX was a full 32-bit processor with a 32-bit bus. The 386 SX
only had a 16-bit bus so it couldn't access more than 16 MB of memory. But back
then, who had that much? They also didn't come with a math coprocessor (that had
to be bought and installed separately). The 486 DX was the one that started
having the FPU on the same die.

Post a reply to this message

Am 03.08.2011 19:40, schrieb s.day:
> clipka<ano### [at] anonymousorg>  wrote:
>>
>> The image was most likely rendered on a 386 DX-40 machine (without
>> mathematical co-processor), using /some/ DOS version of POV-Ray
>> (presumably 2.0). As for the rendering time, I just recall it to have
>> been /awfully/ long.
>
> I thought the DX machines were the ones with the co-processor, my first PC was a
> 386 SX-33 and I remember complaining to the person who built it as I had asked
> for a DX with a co-processor and 4MB of memory. I remember him saying nothing
> has been written to use the co-processors so he did not think there was any
> point in adding one. Also asking me why I needed 4MB of memory, he obviously was
> not aware of POV-Ray.

The 386 processors did not feature an on-die coprocessor yet; AFAIR, the 
SX/DX designation specified the size of the data bus (full 32 bit in the 
DX, multiplexed 2x16 bit in the SX) for this CPU generation.

The first x86 CPU with on-die coprocessor was the 486, with the SX/DX 
designation indeed specifying whether the integrated coprocessor was 
enabled (DX) or disabled (SX, possibly dies with a defective coprocessor 
unit).

(Fun fact: The corresponding chip to upgrade a 486-SX system - the 80487 
- was *not* a coprocessor after all, but essentially a full-fledged 
486-DX CPU instead, which would take over full control of the system, 
disabling the system's mounted 486-SX.)

Post a reply to this message

> Am 03.08.2011 19:40, schrieb s.day:
>> clipka<ano### [at] anonymousorg> wrote:
>>>
>>> The image was most likely rendered on a 386 DX-40 machine (without
>>> mathematical co-processor), using /some/ DOS version of POV-Ray
>>> (presumably 2.0). As for the rendering time, I just recall it to have
>>> been /awfully/ long.
>>
>> I thought the DX machines were the ones with the co-processor, my
>> first PC was a
>> 386 SX-33 and I remember complaining to the person who built it as I
>> had asked
>> for a DX with a co-processor and 4MB of memory. I remember him saying
>> nothing
>> has been written to use the co-processors so he did not think there
>> was any
>> point in adding one. Also asking me why I needed 4MB of memory, he
>> obviously was
>> not aware of POV-Ray.
>
> The 386 processors did not feature an on-die coprocessor yet; AFAIR, the
> SX/DX designation specified the size of the data bus (full 32 bit in the
> DX, multiplexed 2x16 bit in the SX) for this CPU generation.
>
> The first x86 CPU with on-die coprocessor was the 486, with the SX/DX
> designation indeed specifying whether the integrated coprocessor was
> enabled (DX) or disabled (SX, possibly dies with a defective coprocessor
> unit).
>
> (Fun fact: The corresponding chip to upgrade a 486-SX system - the 80487
> - was *not* a coprocessor after all, but essentially a full-fledged
> 486-DX CPU instead, which would take over full control of the system,
> disabling the system's mounted 486-SX.)

The 486 SX also had half sized data bus, thus ths SX designation. 
Essentialy, they made it so that it could be mounted on older, and much 
cheaper, motherboards made for the 386SX.

Post a reply to this message

Alain <aze### [at] qwertyorg> wrote:

> The 486 SX also had half sized data bus, thus ths SX designation.
> Essentialy, they made it so that it could be mounted on older, and much
> cheaper, motherboards made for the 386SX.

Actually, you're thinking of the Pentium Overdrive, which did have a half-sized
bus (32-bits instead of 64-bits for normal Pentiums) and allowed it to be a
plug-in replacement for the 486 DX. The 486 SX had a 32-bit data bus, just like
the 486 DX.

Post a reply to this message