Chambers wrote:
> So I finally built a new machine, meaning I don't have to lease one any 
> more (long story).
> 
> Anyway, I'm happy with it... Phenom 2 X4 at 2.6gHz, 4gB of RAM, and a 
> 1TB hdd :)
> 
> Beta 33 renders the benchmark with 4 threads in 1m58s :) :) :)
> 
> ....Chambers
Congratulations, nice system man, I'm finally upgrading my gear so this 
makes a great reference as well as the others who posted their specs. If 
I were to freely choose my specs would be exactly what you have choosen, 
but since my budget is smaller I think I'll have to saddle for an Phenom 
X3 2.33 Ghz, 4GB DDR2-1066, SEAGATE/MAXTOR 500GB 7200 RPM SATA II HDD.

Post a reply to this message

Chambers schrieb:
> Still, it's my first quad core machine, and I'm excited about it :)

You have reason enough to be. Two months ago, I was still working with a 
P4, and was still so excited about the AMD Phenom 2 I had bought half a 
year ago (as a dedicated Linux/POV-Ray rendering machine :-P) that I was 
about to opt for another Phenom for Windows work, too (for a less 
frustrating modelling experience :-)). Fortunately the guy at the local 
store talked me into buying an i7 instead, but I'm convinced the Phenom 
would have done about as well for quite a few years, too, and I would 
have left me just as excited ;-)

If it wasn't for the modelling and raytracing, I could have lived 
happily with that P4 for another year or two: It's just about the 
computing power I needed for everything else, including software 
development. If it wasn't for games, I'd say a fast P4 (I had already 
upgraded it to the fastest I could get) is as much computing power as 
any sane person could possibly need.

... and we have way over 4 times that power on or under our desks - Gee!

Post a reply to this message

Saul Luizaga wrote:
  > but since my budget is smaller I think I'll have to saddle for an 
Phenom
> X3 2.33 Ghz, 4GB DDR2-1066, SEAGATE/MAXTOR 500GB 7200 RPM SATA II HDD.

I very nearly went for the x3, but since I was planning to upgrade it 
eventually anyway I figured there was no point in buying two processors 
for it.

I might as well get the one I wanted to begin with, and save the money 
by not buying the other :)

...Chambers

Post a reply to this message

TC schrieb:
> I find the results rather shocking - I would not have guessed that switching 
> from 32 to 64 bit would make such a difference.

Neither would I. I had expected the difference to be quite marginal, 
having thought that the only significant benefit of 64-bit was in making 
more memory available for a single application.

> Since you seem to be intimately involved in Pov Ray beta testing,

I actually consider myself more involved in development than in testing, 
but of course I need to test what I have implemented, so yes.

 > maybe you
> can answer the question if the benchmark makes extensive use of memory. 
> Maybe the difference between 3 and 6 GB would account for the difference 
> between 32/64 OS.

Definitely not - I just checked: While rendering the benchmark, POV-Ray 
is utilizing just about 30 MB more than when idle.

I've actually never ever managed to reach even the 2 GB limit on my old 
32-bit machine (well, it didn't have more installed anyway). On the 
64-bit Linux machine, the worst thing I ever observed was about 700 MB 
continuous load with a peak of 800 MB during parsing, IIRC (which was 
actually a problem, becaue I had only 4 GB installed at that time, and 
was running a full-throttle MegaPOV render - i.e. four separate 
instances of POV-Ray. The initial peak memory load kicked the whole 
system into "permanent swapping mode", with the MegaPOV instances 
busying themselves kicking each other (and of course everything else) 
out of physical memory instead of doing anything productive. Operating 
the system was of course no fun anymore until I had managed to pause one 
of the processes. Sustained memory load was low enough that I could 
ultimately resume the fourth process again, but startig them up 
simultaneously was too much.)

Post a reply to this message

Chambers wrote:
> Saul Luizaga wrote:
>  > but since my budget is smaller I think I'll have to saddle for an Phenom
>> X3 2.33 Ghz, 4GB DDR2-1066, SEAGATE/MAXTOR 500GB 7200 RPM SATA II HDD.
> 
> I very nearly went for the x3, but since I was planning to upgrade it 
> eventually anyway I figured there was no point in buying two processors 
> for it.
> 
> I might as well get the one I wanted to begin with, and save the money 
> by not buying the other :)
> 
> ....Chambers
Notice that my Phenom is not Phenom II, just Phenom.

Well if you need the money... but since we POV-Ray you might regret it 
later :)

Post a reply to this message

I have made a little research in www.amd.com and the improvements of 
Phenom II over Phenom are: Socket AM3, DDR3, +2/+4 MB L3 Cache. I think 
I'll make the jump and go for a Phenom II X3 2.8, 6MB L3, 4GB DDR3-1333. 
I can't go for the X4 since is only 2.6 and 4MB L3, each core in the X3 
will run more efficiently than the ones in the X4, and compensate for 
the +2 Ghz of the X4, me thinks.

Post a reply to this message

Another small improvement is +3.9 Ghz in memory bandwidth due to DDR3.

Post a reply to this message

>>> Has anyone ever run the benchmark on an abacus?
>>
>> No, but now I have the urge to create an electromechanical computer 
>> and run the benchmark on that.
> 
> Are you aware how many relays you'll need to even store the rendered 
> image? :P
> 
> Even a device able to compute a b/w image of a RSOCP would be enough of 
> a challenge I guess...

Who said anything about *store*? You could make it output the binary 
data to some kind of storage device.

I would suggest that if you want hardware floating-point support, you're 
still going to need an absurd number of relays though. ;-)

Post a reply to this message

>>> Has anyone ever run the benchmark on an abacus?
>>
>> No, but now I have the urge to create an electromechanical computer 
>> and run the benchmark on that.
>>
>>
> 
> Ah, the Babbage Computational Engine :)

I think you'll find that was mechanical, not electromechanical. :-P

Post a reply to this message

"clipka" <ano### [at] anonymousorg> wrote in message 
news:4a848a81$1@news.povray.org...

> I also heard tell there may also be differences between OS. I'm running XP 
> Pro x64, maybe Vista or Win7 do an even better job at making use of HT. 
> Don't know about this though.

      No, Vista Home is exactly the same time.  :)

      Win7 may be better though.

      ~Steve~

Post a reply to this message

clipka wrote:

> Are you aware how many relays you'll need to even store the rendered 
> image? :P

We have lots of empty space in the warehouse, right now. ;)

> Even a device able to compute a b/w image of a RSOCP would be enough of 
> a challenge I guess...

Definitely. It would be quite fun to build a simple processor like that, 
though

-- 
~Mike

Post a reply to this message

St. wrote:

> 
>       No, Vista Home is exactly the same time.  :)
> 
>       Win7 may be better though.
> 

These are gaming benchmarks, but not much appreciable difference between 
Win7 and Vista:

http://www.tweaktown.com/articles/2869/windows_7_final_oem_vs_windows_vista_vga_performance/index.html

-- 
~Mike

Post a reply to this message

clipka wrote:
> Chambers schrieb:
>> Still, it's my first quad core machine, and I'm excited about it :)
> 
> You have reason enough to be. Two months ago, I was still working with a 
> P4, and was still so excited about the AMD Phenom 2 I had bought half a 
> year ago (as a dedicated Linux/POV-Ray rendering machine :-P) that I was 
> about to opt for another Phenom for Windows work, too (for a less 
> frustrating modelling experience :-)). Fortunately the guy at the local 
> store talked me into buying an i7 instead, but I'm convinced the Phenom 
> would have done about as well for quite a few years, too, and I would 
> have left me just as excited ;-)

No doubt. My major consideration against the i7 was price, but when the 
prices drop, it'll be a less painful upgrade, though I'll have to get 
DDR3 memory.

One of these days I'll switch the OS to the 64 bit version. Retail Win7 
is going to lighten my wallet, though.

> If it wasn't for the modelling and raytracing, I could have lived 
> happily with that P4 for another year or two: It's just about the 
> computing power I needed for everything else, including software 
> development. If it wasn't for games, I'd say a fast P4 (I had already 
> upgraded it to the fastest I could get) is as much computing power as 
> any sane person could possibly need.
> 
> ... and we have way over 4 times that power on or under our desks - Gee!

Yup.

It still surprised me that with the same video chipset, same video 
memory from the same vendor (but PCIe instead of AGP) that gaming 
performance increased as much as it did, since most games don't really 
take advantage of multiple cores, yet!

-- 
~Mike

Post a reply to this message

Clarence1898 wrote:
> 2.4Ghz Intel Quad 6600, 5GB, Vista Home premium (64-bit) - 1m 58s (7m 46s on 1
> cpu)
> 
> I also have Mandriva 2008(32-bit) installed on the Q6600 - 1m 55s (7M 40s on 1
> cpu).
> 
> 
> So on the same hardware 32-bit Linux is very slightly faster than 64-bit Vista.

Interesting, I wonder how much overhead there is in running 32 bit 
applications on 64 bit windows, then?

Makes me want to stick with good ol' 32 bit for a while longer.

-- 
~Mike

Post a reply to this message

>> Even a device able to compute a b/w image of a RSOCP would be enough 
>> of a challenge I guess...
> 
> Definitely. It would be quite fun to build a simple processor like that, 
> though

When I was about 14, I learned all about logic gates, and became 
convinced that one day I would buy a bazillion 7400s and build an actual 
working computer. Hell, when I was at uni I even got as far as buying 
about 16 of them, and some breadboard, wires, etc.

And then I got stumped by the fact that the gates don't appear to do 
what the diagrams suggest they should...

-- 
http://blog.orphi.me.uk/
http://www.zazzle.com/MathematicalOrchid*

Post a reply to this message

Mike Raiford wrote:
> Interesting, I wonder how much overhead there is in running 32 bit 
> applications on 64 bit windows, then?

clipka posted several POV-Ray benchamrk tests about this and 64-64 
(OS-App) was way better than 32-32. I've seen this also in the disease 
research client I run (www.worldcommunitygrid.org), so as soon as I get 
my new PC I'm switching to 64-bits (WinXP x64, no Vista nor Win7). My 
main interest is the disease research client, I want to do more, alot 
more and 64-bit integers does one of the tricks to make this client go 
way faster.

Post a reply to this message

Orchid XP v8 schrieb:
> And then I got stumped by the fact that the gates don't appear to do 
> what the diagrams suggest they should...

In what way?

Post a reply to this message

Orchid XP v8 wrote:
> And then I got stumped by the fact that the gates don't appear to do 
> what the diagrams suggest they should...

Which brings us to the debate about whether information is intrinsic or 
an added characteristic.

That is, does the diagram actually say anything, or do you invent the 
meaning yourself? :)

...Chambers

Post a reply to this message

>> And then I got stumped by the fact that the gates don't appear to do 
>> what the diagrams suggest they should...
> 
> In what way?

The 7400 is a quad 2-input NAND gate. In other words, it's a box with 
four NAND gates in it.

What's supposed to happen is that you connect pins 1 and 14 to a power 
source, and the remaining pins are grouped into 3s. Each group is the 
connections for a single NAND gate. When the inputs aren't connected, 
the output pin is at logic high. When you connect both of the input pins 
to the (+) rail, the output should go low.

...except this doesn't seem to happen. I swear to God I built circuits 
like this when I was a kid, and it worked. But when I tried it as a 
teenager, it wouldn't work for toffee. No idea why.

-- 
http://blog.orphi.me.uk/
http://www.zazzle.com/MathematicalOrchid*

Post a reply to this message

On Sat, 15 Aug 2009 07:21:00 +0100, Orchid XP v8 <voi### [at] devnull> wrote:

>>> And then I got stumped by the fact that the gates don't appear to do 
>>> what the diagrams suggest they should...
>> 
>> In what way?
>
>The 7400 is a quad 2-input NAND gate. In other words, it's a box with 
>four NAND gates in it.
>
>What's supposed to happen is that you connect pins 1 and 14 to a power 
>source, and the remaining pins are grouped into 3s. Each group is the 
>connections for a single NAND gate. When the inputs aren't connected, 
>the output pin is at logic high. When you connect both of the input pins 
>to the (+) rail, the output should go low.
>
>...except this doesn't seem to happen. I swear to God I built circuits 
>like this when I was a kid, and it worked. But when I tried it as a 
>teenager, it wouldn't work for toffee. No idea why.


Good grief Andrew, 7400s, did you rob a museum? Do you know the power
consumption of TTL compared to CMOS? 
Besides pin 14 being Vcc and pin 7 ground. You cannot assume that inputs will
float high you really should drive them high or low.

-- 

Regards
     Stephen

Post a reply to this message

>> The 7400 is a quad 2-input NAND gate. In other words, it's a box with 
>> four NAND gates in it.
>>
>> What's supposed to happen is that you connect pins 1 and 14 to a power 
>> source, and the remaining pins are grouped into 3s. Each group is the 
>> connections for a single NAND gate. When the inputs aren't connected, 
>> the output pin is at logic high. When you connect both of the input pins 
>> to the (+) rail, the output should go low.
>>
>> ...except this doesn't seem to happen. I swear to God I built circuits 
>> like this when I was a kid, and it worked. But when I tried it as a 
>> teenager, it wouldn't work for toffee. No idea why.
> 
> 
> Good grief Andrew, 7400s, did you rob a museum?

Maplin.

IIRC, they charge 21.22p per chip. (Presumably the 0.22p only makes a 
difference if you buy hundreds of them...)

> Do you know the power consumption of TTL compared to CMOS?

TTL has the advantage that it doesn't break if you touch it. ;-)

> Besides pin 14 being Vcc and pin 7 ground.

OK. I was going from memory there.

> You cannot assume that inputs will
> float high you really should drive them high or low.

...?

-- 
http://blog.orphi.me.uk/
http://www.zazzle.com/MathematicalOrchid*

Post a reply to this message

On Sat, 15 Aug 2009 09:10:26 +0100, Orchid XP v8 <voi### [at] devnull> wrote:

 
>> Good grief Andrew, 7400s, did you rob a museum?
>
>Maplin.
>

When I worked for Motorola way back in the mid 70's. Maplin and RS were sold the
ICs that failed all the tests except the functional ones. So I would not trust
the quality very much. 

>IIRC, they charge 21.22p per chip. (Presumably the 0.22p only makes a 
>difference if you buy hundreds of them...)
>

I suppose I was spoilt as I always worked for a company that would let me have
them free. FWIW It was considered an educational thing to allow engineers some
time and materials for home jobs.

>> Do you know the power consumption of TTL compared to CMOS?
>
>TTL has the advantage that it doesn't break if you touch it. ;-)
>

Fair point although I never had much problem with ESD and CMOS as long as you
earth yourself. 

>> Besides pin 14 being Vcc and pin 7 ground.
>
>OK. I was going from memory there.
>

I guessed as much.

>> You cannot assume that inputs will
>> float high you really should drive them high or low.
>
>...?

IIRC I think the specs said that TTL inputs would be at a logical 1 if there was
no connection. (Internal resistor to Vcc) 
This was not the case with CMOS, so I was wrong saying that TTL had to be driven
either high or low. But it is good practice to drive your inputs in case there
are things in your physical circuit that might affect what you think will
happen. Such as dry joints, solder bridges, RF pickup etc.
-- 

Regards
     Stephen

Post a reply to this message

>>> Good grief Andrew, 7400s, did you rob a museum?
>> Maplin.
> 
> When I worked for Motorola way back in the mid 70's. Maplin and RS were sold the
> ICs that failed all the tests except the functional ones. So I would not trust
> the quality very much. 

I'm told 10% resistors show a curios bimodel resistance distribution 
too. ;-)

BTW... What tests are there *other* than functional ones?!

> FWIW It was considered an educational thing to allow engineers some
> time and materials for home jobs.

Indeed. I'm told Cadbury let their employees eat as much chocolate as 
they want. Not so much for educational purposes, but because the amount 
of chocolate that one employee can eat without being horrifyingly sick 
is insignificant compared to the amount of chocolate they make per day...

>>> Do you know the power consumption of TTL compared to CMOS?
>> TTL has the advantage that it doesn't break if you touch it. ;-)
> 
> Fair point although I never had much problem with ESD and CMOS as long as you
> earth yourself. 

What voltage does CMOS use anyway? IIRC, TTL can be driven with a 
triplet of 1.5 V batteries.

>> OK. I was going from memory there.
> 
> I guessed as much.

It's been a while since I did all this.

> IIRC I think the specs said that TTL inputs would be at a logical 1 if there was
> no connection. (Internal resistor to Vcc) 

Oh.

Well that would be a problem then. I was assuming it would be logic 0. o_O

> This was not the case with CMOS, so I was wrong saying that TTL had to be driven
> either high or low. But it is good practice to drive your inputs in case there
> are things in your physical circuit that might affect what you think will
> happen. Such as dry joints, solder bridges, RF pickup etc.

So connecting to the negative rail should give me 0, and connecting to 
the positive rail should give 1?

-- 
http://blog.orphi.me.uk/
http://www.zazzle.com/MathematicalOrchid*

Post a reply to this message

On Sat, 15 Aug 2009 12:37:30 +0100, Orchid XP v8 <voi### [at] devnull> wrote:

>>>> Good grief Andrew, 7400s, did you rob a museum?
>>> Maplin.
>> 
>> When I worked for Motorola way back in the mid 70's. Maplin and RS were sold the
>> ICs that failed all the tests except the functional ones. So I would not trust
>> the quality very much. 
>
>I'm told 10% resistors show a curios bimodel resistance distribution 
>too. ;-)
>

It would not surprise me but I could not say for sure,

>BTW... What tests are there *other* than functional ones?!
>

Well it was quite a while ago and I assume that they will have changed.
IIRC: Rise and fall times of the output, propagation delay of the gates, input
and output voltage range, operational temperature and humidly. There was a
centrifuge test up to 200 g for mil spec and a test where the ICs were dipped in
liquid helium and tested to see if there were any leaks on the encapsulation.


>> FWIW It was considered an educational thing to allow engineers some
>> time and materials for home jobs.
>
>Indeed. I'm told Cadbury let their employees eat as much chocolate as 
>they want. Not so much for educational purposes, but because the amount 
>of chocolate that one employee can eat without being horrifyingly sick 
>is insignificant compared to the amount of chocolate they make per day...
>

I believe that too.

>>>> Do you know the power consumption of TTL compared to CMOS?
>>> TTL has the advantage that it doesn't break if you touch it. ;-)
>> 
>> Fair point although I never had much problem with ESD and CMOS as long as you
>> earth yourself. 
>
>What voltage does CMOS use anyway? IIRC, TTL can be driven with a 
>triplet of 1.5 V batteries.
>

Why don't you look it up as I can't remember,

>>> OK. I was going from memory there.
>> 
>> I guessed as much.
>
>It's been a while since I did all this.
>
>> IIRC I think the specs said that TTL inputs would be at a logical 1 if there was
>> no connection. (Internal resistor to Vcc) 
>
>Oh.
>
>Well that would be a problem then. I was assuming it would be logic 0. o_O
>

:-)

>> This was not the case with CMOS, so I was wrong saying that TTL had to be driven
>> either high or low. But it is good practice to drive your inputs in case there
>> are things in your physical circuit that might affect what you think will
>> happen. Such as dry joints, solder bridges, RF pickup etc.
>
>So connecting to the negative rail should give me 0, and connecting to 
>the positive rail should give 1?

Yes, by George you've got it!

:)
-- 

Regards
     Stephen

Post a reply to this message

On 15-8-2009 13:37, Orchid XP v8 wrote:
>>>> Good grief Andrew, 7400s, did you rob a museum?

> 
>> IIRC I think the specs said that TTL inputs would be at a logical 1 if 
>> there was
>> no connection. (Internal resistor to Vcc) 
> 
> Oh.
> 
> Well that would be a problem then. I was assuming it would be logic 0. o_O

Starting middle of September I'll be teaching digital technology in the 
first year of the University of Applied Science in Amsterdam. If you 
happen to be around, just drop in. I think I will do the 7400 in the 
week of Sept 21. ;)

As others have said, the original TTL series by Texas Instruments acted 
as if a one was present if left open. All other gates should do the same 
because of backwards compatibility. That includes the CMOS pin 
compatible replacements. I have actually seen rare default 0 gates but I 
would consider those faulty.

When designing the rule is not to leave inputs open. Not even for gates 
you don't use. One reason is that they could pick up interference which 
could result in unpredictable behaviour. For CMOS gates it could also 
result in useless power consumption. In any case the inputs are often 
next to one another like in the 7400. It is just as easy to wire them 
together (connecting them to ground (for NOR gates) or the positive rail 
(for NAND gates) is often more complicated).

Oh, and if you want to build your own computer, I'd suggest using also 
other components apart from the 7400.

Post a reply to this message

On Sat, 15 Aug 2009 22:08:45 +0200, andrel <a_l### [at] hotmailcom> wrote:

>Oh, and if you want to build your own computer, I'd suggest using also 
>other components apart from the 7400.

Why? :P

Maybe some JK and D flip-flops
-- 

Regards
     Stephen

Post a reply to this message

Orchid XP v8 schrieb:
>>> And then I got stumped by the fact that the gates don't appear to do 
>>> what the diagrams suggest they should...
>>
>> In what way?
> 
> The 7400 is a quad 2-input NAND gate. In other words, it's a box with 
> four NAND gates in it.
> 
> What's supposed to happen is that you connect pins 1 and 14 to a power 
> source, and the remaining pins are grouped into 3s. Each group is the 
> connections for a single NAND gate. When the inputs aren't connected, 
> the output pin is at logic high. When you connect both of the input pins 
> to the (+) rail, the output should go low.
> 
> ...except this doesn't seem to happen. I swear to God I built circuits 
> like this when I was a kid, and it worked. But when I tried it as a 
> teenager, it wouldn't work for toffee. No idea why.

Maybe some other technology? While classic TTL gates (74xx) as well as 
Low Power Schottky variants (74LSxx) would internally pull up 
unconnected inputs, CMOS gates (74HCxx and 74HCTxx) would lack a pull-up 
resistor, causing unconnedted inputs to float erratically.

To work properly, all input pins of a CMOS IC /must/ therefore be 
actively pulled to either VCC or GND (a resistor is not required in that 
case, as CMOS ICs draw virtually no current on static input pins, so you 
can directly connect them to VCC or GND). It is even highly recommended 
to even connect all input pins of unused gates in an IC package.

Post a reply to this message

Stephen schrieb:
>> What voltage does CMOS use anyway? IIRC, TTL can be driven with a 
>> triplet of 1.5 V batteries.
> 
> Why don't you look it up as I can't remember,

74HCxx: 2V to 6V (i.e. a doublet or triplet of 1.5 V batteries will do 
fine, even when quite depleted already)

74HCTxx: 5V, +/- 0.5V (i.e. a triplet will probably do quite well if 
still good)

Power consumption is virtually zero at low frequencies.

Post a reply to this message

>>> IIRC I think the specs said that TTL inputs would be at a logical 1 
>>> if there was no connection. (Internal resistor to Vcc) 
>>
>> Oh.
>>
>> Well that would be a problem then. I was assuming it would be logic 0. 
>> o_O
> 
> Starting middle of September I'll be teaching digital technology in the 
> first year of the University of Applied Science in Amsterdam. If you 
> happen to be around, just drop in. I think I will do the 7400 in the 
> week of Sept 21. ;)

Heh. I doubt it - but nice idea. ;-)

> As others have said, the original TTL series by Texas Instruments acted 
> as if a one was present if left open. All other gates should do the same 
> because of backwards compatibility. That includes the CMOS pin 
> compatible replacements. I have actually seen rare default 0 gates but I 
> would consider those faulty.
> 
> When designing the rule is not to leave inputs open. Not even for gates 
> you don't use. One reason is that they could pick up interference which 
> could result in unpredictable behaviour. For CMOS gates it could also 
> result in useless power consumption. In any case the inputs are often 
> next to one another like in the 7400. It is just as easy to wire them 
> together (connecting them to ground (for NOR gates) or the positive rail 
> (for NAND gates) is often more complicated).

Mmm. This I was not aware of...

> Oh, and if you want to build your own computer, I'd suggest using also 
> other components apart from the 7400.

OOC, exactly how much space do you recon it would take to build 64KB out 
of discrete NAND gates?

-- 
http://blog.orphi.me.uk/
http://www.zazzle.com/MathematicalOrchid*

Post a reply to this message

>> The 7400 is a quad 2-input NAND gate. In other words, it's a box with 
>> four NAND gates in it.
>>
>> What's supposed to happen is that you connect pins 1 and 14 to a power 
>> source, and the remaining pins are grouped into 3s. Each group is the 
>> connections for a single NAND gate. When the inputs aren't connected, 
>> the output pin is at logic high. When you connect both of the input pins 
>> to the (+) rail, the output should go low.
> 
> Good grief Andrew, 7400s, did you rob a museum? Do you know the power
> consumption of TTL compared to CMOS? 
> Besides pin 14 being Vcc and pin 7 ground. You cannot assume that inputs will
> float high you really should drive them high or low.

I just looked up the wiring diagram.

It appears I had the right pins, they're just not numbered in the order 
I was expecting.

-- 
http://blog.orphi.me.uk/
http://www.zazzle.com/MathematicalOrchid*

Post a reply to this message

On Sun, 16 Aug 2009 01:06:12 +0200, clipka <ano### [at] anonymousorg> wrote:

>Stephen schrieb:
>>> What voltage does CMOS use anyway? IIRC, TTL can be driven with a 
>>> triplet of 1.5 V batteries.
>> 
>> Why don't you look it up as I can't remember,
>
>74HCxx: 2V to 6V (i.e. a doublet or triplet of 1.5 V batteries will do 
>fine, even when quite depleted already)
>
>74HCTxx: 5V, +/- 0.5V (i.e. a triplet will probably do quite well if 
>still good)
>
>Power consumption is virtually zero at low frequencies.

An example: Around 1975 I made a digital clock out of TTL and the power
consumption was 20 amps. About 1977 I made another using CMOS, it ran for years
on a PP3 battery.
-- 

Regards
     Stephen

Post a reply to this message

On Sun, 16 Aug 2009 09:16:51 +0100, Orchid XP v8 <voi### [at] devnull> wrote:

>OOC, exactly how much space do you recon it would take to build 64KB out 
>of discrete NAND gates?

For memory you could use a D type flip-flop which can be made using 4 NAND gates
and an inverting gate. It would be better to use a Master-slave D type flip-flop
which would double the components. So for each *bit* of memory you would need 2
X 7400s and 1/3 X 7404. or a 7474 or a 7479.

What size of word are you thinking of? IIRC the typical power consumption of TTL
is about 250 ma. You will need a robust power supply.
-- 

Regards
     Stephen

Post a reply to this message

>> OOC, exactly how much space do you recon it would take to build 64KB out 
>> of discrete NAND gates?
> 
> For memory you could use a D type flip-flop which can be made using 4 NAND gates
> and an inverting gate. It would be better to use a Master-slave D type flip-flop
> which would double the components. So for each *bit* of memory you would need 2
> X 7400s and 1/3 X 7404. or a 7474 or a 7479.
> 
> What size of word are you thinking of? IIRC the typical power consumption of TTL
> is about 250 ma. You will need a robust power supply.

Of course, I have no serious *intention* of making more than about a 
dozen bytes of RAM using only NAND gates. ;-) But I'm aware that way 
back in prehistoric times, they used to do it this way.

No wonder computers cost more that housing estates - they were *bigger* 
than housing estates! :-D

-- 
http://blog.orphi.me.uk/
http://www.zazzle.com/MathematicalOrchid*

Post a reply to this message

triple_r wrote:
> a computer that gets upset when you make programming errors!  There must be
> room for an "In Soviet Russia..." joke somewhere around here.

In Soviet Russia, the computer hacks you.

>  - Ricky

-Aero

Post a reply to this message

On Sun, 16 Aug 2009 14:09:32 +0100, Orchid XP v8 <voi### [at] devnull> wrote:

>
>Of course, I have no serious *intention* of making more than about a 
>dozen bytes of RAM using only NAND gates. ;-)

Surprise me :D

> But I'm aware that way back in prehistoric times, they used to do it this way.
>

Back in prehistoric days, at least when I started working in computing, RAM was
ferrite-core memory. Literally ferrite or ceramic toroids threaded with wires to
store information via the polarity of the magnetic fields.
http://en.wikipedia.org/wiki/Core_memory
 

>No wonder computers cost more that housing estates - they were *bigger* 
>than housing estates! :-D

It was the air conditioning that took up the room :)
-- 

Regards
     Stephen

Post a reply to this message

On 16-8-2009 15:09, Orchid XP v8 wrote:
>>> OOC, exactly how much space do you recon it would take to build 64KB 
>>> out of discrete NAND gates?
>>
>> For memory you could use a D type flip-flop which can be made using 4 
>> NAND gates
>> and an inverting gate. It would be better to use a Master-slave D type 
>> flip-flop
>> which would double the components. So for each *bit* of memory you 
>> would need 2
>> X 7400s and 1/3 X 7404. or a 7474 or a 7479.
>>
>> What size of word are you thinking of? 

let's assume 8 bit and that we need about as much logic to select the 
bits. That'll be a breadboard if about 25 by 25 meter (modulo arithmetic 
errors), so not too bad.

> IIRC the typical power 
>> consumption of TTL
>> is about 250 ma. You will need a robust power supply.

Cooling was a more difficult problem than the logical design of 
supercomputers.

> Of course, I have no serious *intention* of making more than about a 
> dozen bytes of RAM using only NAND gates. ;-) But I'm aware that way 
> back in prehistoric times, they used to do it this way.

I don't think they did it with 7400s. The description of the ENIAC 
suggests that they did something similar, though I would not be 
surprised if they used a slightly different design. By that time the 
tubes were much more versatile than simple transistors.

A popular one afrom the beginning of mainframe computers was core memory 
(see e.g. http://www.columbia.edu/acis/history/core.html).

> No wonder computers cost more that housing estates - they were *bigger* 
> than housing estates! :-D

That was only in a very short period of time. Aside the analog computers 
then were just as big.

Post a reply to this message

On Sun, 16 Aug 2009 16:55:46 +0200, andrel <a_l### [at] hotmailcom> wrote:

>A popular one afrom the beginning of mainframe computers was core memory 
>(see e.g. http://www.columbia.edu/acis/history/core.html).

Just wandering down memory lane I looked up the Honeywell H316 as it was the
first computer that I worked on. At one time marketed as "The Honeywell Kitchen
Computer" LOL http://en.wikipedia.org/wiki/Honeywell_316

http://www.bitsavers.org/pdf/honeywell/series16/h316/70130072179A_RealTimeClk_May69.pdf

I found a site with lots of documents even circuit diagrams
http://www.bitsavers.org/pdf/honeywell/series16/h316/

36 Meg pdf
http://www.bitsavers.org/pdf/honeywell/series16/h316/70130072166AV_316_ModulesAndParts_Nov74.pdf

Bookmark 2-07 gives the "Principles of Magnetic Core Operation"

-- 

Regards
     Stephen

Post a reply to this message

Orchid XP v8 schrieb:
> It appears I had the right pins, they're just not numbered in the order 
> I was expecting.

That /might/ lead to incorrectly wired circuitry :-P

Post a reply to this message

>> It appears I had the right pins, they're just not numbered in the 
>> order I was expecting.
> 
> That /might/ lead to incorrectly wired circuitry :-P

The point being, it *is* the two corner pins, as I expected. It's just 
that those corners aren't numbered 1 and 14 for some reason.

-- 
http://blog.orphi.me.uk/
http://www.zazzle.com/MathematicalOrchid*

Post a reply to this message

Since I'm currently using the onboard video adapter (Radeon 3200, they 
call it), I wasn't expecting to play many games until I bought a new 
video card.

However, I'm quite impressed by the performance of it.  Company of 
Heroes runs fine (with options turned down, of course ;) ), and HL2 ran 
well enough that I decided to get Ep 1&2 (they're on sale, and I've 
never gotten around to playing them).

Ooh... I should install Civ IV again!  That should run just fine :)

(As if I needed more ways to waste my time ;) )

...Chambers

Post a reply to this message

Orchid XP v8 schrieb:
> The point being, it *is* the two corner pins, as I expected. It's just 
> that those corners aren't numbered 1 and 14 for some reason.

You're talking about the power pins?

They are typically found at the opposite corners of the chip, "bottom 
left" being GND and "top right" being VCC. Not only for 74xx, but for a 
great number of stuff commonly found in DIL packages (memory ICs, 
microcontrollers and what-have-you-nots).

The numbering seems to have evolved independently from this, following a 
"U" shape starting at "top left". This is /always/ the case for DIL 
packages (with problems arising only if pins are completely left out; I 
think they're counted as if they were there).

Indeed this, combined, is not very intuitive.

Post a reply to this message

On 16-8-2009 19:34, Orchid XP v8 wrote:
>>> It appears I had the right pins, they're just not numbered in the 
>>> order I was expecting.
>>
>> That /might/ lead to incorrectly wired circuitry :-P
> 
> The point being, it *is* the two corner pins, as I expected. It's just 
> that those corners aren't numbered 1 and 14 for some reason.

Let me guess: valves

Post a reply to this message

On 16-8-2009 17:27, Stephen wrote:
> On Sun, 16 Aug 2009 16:55:46 +0200, andrel <a_l### [at] hotmailcom> wrote:
> 
>> A popular one afrom the beginning of mainframe computers was core memory 
>> (see e.g. http://www.columbia.edu/acis/history/core.html).
> 
> Just wandering down memory lane I looked up the Honeywell H316 as it was the
> first computer that I worked on. At one time marketed as "The Honeywell Kitchen
> Computer" LOL http://en.wikipedia.org/wiki/Honeywell_316

Seems you are a few years before into computers. I started programming 
on (I think) a TRS80 in 1978 or 1979. Although I might have encountered 
a computer before that at my summer holiday job, which had something to 
do with punched cards. My first serious mainframe was a CDC Cyber 
stationed at SARA, the computer facility in Amsterdam in 1982.
I just found some stuff from that era. Yip does bring back memories.

Post a reply to this message

On Sun, 16 Aug 2009 21:55:11 +0200, andrel <a_l### [at] hotmailcom> wrote:

>Seems you are a few years before into computers.

My first real job was with Honeywell, in 1969. Where I tested PCBs for the H-416
and H-316. The first computer I ever saw was a valve analog one at Glasgow
University a couple of years earlier. I spent about 10 years working in computer
maintenance in factories before working offshore. Which was a very different
life :)
Now I skive around working in SAP configuring the Plant Maintenance and Customer
Service modules.
-- 

Regards
     Stephen

Post a reply to this message

On 16-8-2009 22:24, Stephen wrote:
> On Sun, 16 Aug 2009 21:55:11 +0200, andrel <a_l### [at] hotmailcom> wrote:
> 
>> Seems you are a few years before into computers.
> 
> My first real job was with Honeywell, in 1969. Where I tested PCBs for the H-416
> and H-316. The first computer I ever saw was a valve analog one at Glasgow
> University a couple of years earlier. I spent about 10 years working in computer
> maintenance in factories before working offshore. Which was a very different
> life :)

Hey, another off-shore guy. Coincidence? I guess so.

> Now I skive around working in SAP configuring the Plant Maintenance and Customer
> Service modules.

Off to Google... Ok, SAP is that sort of company. Having plant and sap 
in one sentence might suggest something else. Just kidding, I simply had 
not heard of it. And I assume skive means something with a lot of 
travelling.

Post a reply to this message

On Sun, 16 Aug 2009 22:34:00 +0200, andrel <a_l### [at] hotmailcom> wrote:


>Hey, another off-shore guy. Coincidence? I guess so.
>

You too?

Shay works offshore in the Mexican Gulf IIRC

>> Now I skive around working in SAP configuring the Plant Maintenance and Customer
>> Service modules.
>
>Off to Google... Ok, SAP is that sort of company. Having plant and sap 
>in one sentence might suggest something else. Just kidding, 

LOL

>I simply had 
>not heard of it. And I assume skive means something with a lot of 
>travelling.

http://en.wikipedia.org/wiki/Skive

-- 

Regards
     Stephen

Post a reply to this message

On 16-8-2009 22:45, Stephen wrote:
> On Sun, 16 Aug 2009 22:34:00 +0200, andrel <a_l### [at] hotmailcom> wrote:
> 
> 
>> Hey, another off-shore guy. Coincidence? I guess so.
>>
> 
> You too?

No, referring to the Shay of the line below.

> Shay works offshore in the Mexican Gulf IIRC

I think he is grounded again.

> 
>>> Now I skive around working in SAP configuring the Plant Maintenance and Customer
>>> Service modules.
>> And I assume skive means something with a lot of 
>> travelling.
> 
> http://en.wikipedia.org/wiki/Skive

Ok, I have to work on my Brittish slang.

Post a reply to this message

On Sun, 16 Aug 2009 22:56:56 +0200, andrel <a_l### [at] hotmailcom> wrote:

>No, referring to the Shay of the line below.
>
>> Shay works offshore in the Mexican Gulf IIRC
>
>I think he is grounded again.
>

???

>> 
>>>> Now I skive around working in SAP configuring the Plant Maintenance and Customer
>>>> Service modules.
>>> And I assume skive means something with a lot of 
>>> travelling.
>> 
>> http://en.wikipedia.org/wiki/Skive
>
>Ok, I have to work on my Brittish slang.

I'll help where I can :P
-- 

Regards
     Stephen

Post a reply to this message

On 16-8-2009 23:02, Stephen wrote:
> On Sun, 16 Aug 2009 22:56:56 +0200, andrel <a_l### [at] hotmailcom> wrote:
> 
>> No, referring to the Shay of the line below.
>>
>>> Shay works offshore in the Mexican Gulf IIRC
>> I think he is grounded again.
>>
> 
> ???

To show of my lack of knowledge I decided to use an expression that 
probably would not mean 'he has a job on land now' not even for offshore 
people. I think I succeeded.

>>>>> Now I skive around working in SAP configuring the Plant Maintenance and Customer
>>>>> Service modules.
>>>> And I assume skive means something with a lot of 
>>>> travelling.
>>> http://en.wikipedia.org/wiki/Skive
>> Ok, I have to work on my Brittish slang.
> 
> I'll help where I can :P

Post a reply to this message

On Mon, 17 Aug 2009 00:34:57 +0200, andrel <a_l### [at] hotmailcom> wrote:

>>>> Shay works offshore in the Mexican Gulf IIRC
>>> I think he is grounded again.
>>>
>> 
>> ???
>
>To show of my lack of knowledge I decided to use an expression that 
>probably would not mean 'he has a job on land now' not even for offshore 
>people. I think I succeeded.

"Grounded" specifically means not allowed to by an authority, amongst other
things. A pilot or plane can be grounded by weather or for safety reasons. A
child may be grounded by its parents as a punishment.   
If Shay were working onshore, in Britain, we would say that he is "on the
beach". But I'm sure he will let us know when he next pops in.
-- 

Regards
     Stephen

Post a reply to this message