Strangely high serial number for Fidelity SC9

[Larry]

Guys, I can tell that not all SC9B's had the daughter board. Here is my other SC9B main board.
https://flic.kr/p/UE471t
serial# 32664328
It is definitely an SC9B, you can tell the higher pitch startup beeps compared
to the "A". Also, it passes the "1h4...b6" test.
Note the slightly messy arrangement, with two chips piggy backing
each other. This is not a modified main board, I've seen another one
the same.
L.

[Larry]

Regarding the CB16 (serial#22391083) module, if you look closely at
the picture you will see slight bumps in the decal. Under these bumps
are the two screws holding the unit together. It can only be pulled
apart by damaging the decal. Sorry, but I'm not willing to do that. I'm
sure you will understand.
https://flic.kr/p/TC3Gdk

[yoyo_chessboard]

Hi Larry

about the startup beep and more generally about sound produced by the boards.

the sound routine is a hard-coded routine with a fixed number of instructions that produces a predefined sound at a given frequency.
when you change the frequency the sound is different :

sharper when you have a faster clock
louder when you have a slower clock
so your observation is right but only when we have the same binary (which is your case).

so for a given binary/model , same sounds means same clock speed and sharper sounds means higher clockrate.
that is a first trick

so if your two SC9x have same ROM ID, that means that the second one (the one with the IC sandwich) has a higher clockrate than the previous one.

second trick
but you can't compare sound from two different binaries : for exemple excellence and SC9 , or x with y, or even two variants in the excellence family (EP12, 6080, 6080B, par excellence,...).

because the sound routine is different from one binary to the other.
and the length of the loop that generates the sounds is different.

[yoyo_chessboard]

Guys, I can tell that not all SC9B's had the daughter board. Here is my other SC9B main board.
https://flic.kr/p/UE471t
serial# 32664328
It is definitely an SC9B, you can tell the higher pitch startup beeps compared
to the "A". Also, it passes the "1h4...b6" test.
Note the slightly messy arrangement, with two chips piggy backing
each other. This is not a modified main board, I've seen another one
the same.
L.

HI Larry

your board is interesting

serial 3266 suggest that it is from September 1983, say 23 or 24 september.
on the other side, we know (or think we know) that SC9B were introduced in october 1983.

so you might have one of the very first SC9B if ROM IDs matches.

could you look at the ROM_ IDs and datecode ?
are there the same as in your other board ?

for datecode it can't be the same as your second board's ROMs have datecode of november 1983 (week 47 and 46)
and mike's were 43/44 IIRC.

maybe a new set of photos would be great if board is still opened.

[Mike Watters]

Hi Guys

For what it is worth here is a more complete picture of my 9B SN 33295753 showing the daughter board.



Without knowing much about the 9B I would have thought that the daughter board is typical Fidelity during that period.
(1) Keep costs down by making large numbers of pcbs, cases, ROMs etc
(2) Use the components you have to incorporate into upgraded chess computers or mix-and-match special editions when the original design is beginning to get less competitive.
Presumably the daughter board is the most cost effective way of upgrading the 9A?

All the best
Mike

[Larry]

Yes Mike, I thought the exact same thing. If mine was an early 9B
then the company would have it way more expedient to have
daughter boards flown in from China at 40 cents each and just push-fit
them, rather than frig around soldering like you saw on mine. Even
considering they were the days before surface mount boards.
L

[yoyo_chessboard]

may be one of our difficulty is a problem of definition :

what is an SC9A and what is an SC9B:

for me :
a SC9A is having the following ROMs (extract from MAME emulation):

ROM_START( fscc9 )
ROM_LOAD("101-1034a01", 0xc000, 0x2000, CRC(b845c458) SHA1(d3fda65dbd9fae44fa4b93f8207839d8fa0c367a) ) // HN48364P
ROM_LOAD("101-1034b02", 0xe000, 0x2000, CRC(cbaf97d7) SHA1(7ed8e68bb74713d9e2ff1d9c037012320b7bfcbf) ) // "


and a SC9B :

ROM_START( fscc9b ) // this one came from an overclocked board, let's assume the roms were unmodified
ROM_LOAD("101-1034b01", 0xc000, 0x2000, CRC(65288753) SHA1(651f5ca5969ddd72a20cbebdec2de83c4bf10650) )
ROM_LOAD("101-1034c02", 0xe000, 0x2000, CRC(238b092f) SHA1(7ddffc6dba822aee9d8ad6815b23024ed5cdfd26) )


whatever clockrate they have .
for example my 3 SC9A (according to the previous definition) are: 1.47, 1.47 and 1.8 Mhz)
for example my 1.8 Mhz SC9A is obsiously faster than my 1.4 Mhz.
and when i modified my SC9A, it was faster.
but it is still a SC9A engine.

do you agree with that definition of SC9A/B ?

many sellers on Ebay refers to some speed test to claim their SC9 is a B version or refer to schach computer infos, even if their serial number is too small , lower than 332xxxxxx).

larry's and mike are SC9B and all 4069xxxx also.

what do you think of that definition ?

[Steve B]

Not following the technical specs, serial numbers and Mother Board photos in his thread but I am fairly confident that I have a bone fide SC9A and SC9B
i bought the SC9B years ago ..when first released for sale..from a retailer..
it was sold to me as an upgraded SC9(what we now think of as a "B")

as has been mentioned before here in different threads ....

in my A.. at the 5 second level... it will answer 1.h4 with either 1..e5 or 1..d5
only those two moves
with my "B" it replies to 1.h4 with 1..b6 only

does any one have what they think is an SC9A that replies to 1.h4 with 1..b6(at 5 seconds)??
OR CONVERSLY
does anyone have what they think is a "B" that replies to 1.h4 with either 1..e5 or 1..d5??

If not ..then a simple working definition of an SC9B is that it will reply to 1.h4 with 1..b6

Won't Pry Open My Computer...No How No Way Regards
Steve

[spacious_mind]

My opinion on this.

A program has to prove that it has a different play behavior in given situations to become a version change. ie.. Super Expert A, B, C.

If any changes are as a result of speed increases then in my opinion it should be labeled as for example SC9-1 MHz, SC9-2 MHz etc. since the program itself remains unchanged.

Steve's Test would be a good indication for version changes, however even Steve's test leaves room for doubt, because the possibility still remains that the Version B because of its double or more speed increase will change its move choice because it has the ability to search the position deeper.

Therefore assuming "B" were to be twice as fast then in any fixed position tests to compare it accurately to an "A" you would also have to allow "B" half the time for these fixed position tests.

Best regards

[Steve B]

My opinion on this.

A program has to prove that it has a different play behavior in given situations to become a version change. ie.. Super Expert A, B, C.

If any changes are as a result of speed increases then in my opinion it should be labeled as for example SC9-1 MHz, SC9-2 MHz etc. since the program itself remains unchanged.

Steve's Test would be a good indication for version changes, however even Steve's test leaves room for doubt, because the possibility still remains that the Version B because of its double or more speed increase will change its move choice because it has the ability to search the position deeper.

Therefore assuming "B" were to be twice as fast then in any fixed position tests to compare it accurately to an "A" you would also have to allow "B" half the time for these fixed position tests.

Best regards

Hi Nick
IIRC the move was a book move reply for the SC9B so the time makes no difference
however its not hard to check this
even though the SC9B is not running twice as fast as the SC9A I will give the SC9A 10 seconds to move (vs the SC9B's 5 )and see what it comes up with after 1.h4
my guess us the SC9A will still play either 1..e5 or 1..d5

B6 Regards
Steve

[spacious_mind]

My opinion on this.

A program has to prove that it has a different play behavior in given situations to become a version change. ie.. Super Expert A, B, C.

If any changes are as a result of speed increases then in my opinion it should be labeled as for example SC9-1 MHz, SC9-2 MHz etc. since the program itself remains unchanged.

Steve's Test would be a good indication for version changes, however even Steve's test leaves room for doubt, because the possibility still remains that the Version B because of its double or more speed increase will change its move choice because it has the ability to search the position deeper.

Therefore assuming "B" were to be twice as fast then in any fixed position tests to compare it accurately to an "A" you would also have to allow "B" half the time for these fixed position tests.

Best regards

Hi Nick
IIRC the moves were book move replies so the time makes no difference
however its not hard to test this
will give the SC9A 10 seconds to move (vs the SC9B's 5 )and see what it comers up with after 1.h4
even though the SC9B is not running twice as fast as the SC9A my guess us the SC9A will still play either 1..e5 or 1..d5

B6 Regards
Steve

Thanks, for clarifying Steve, just like you I avoid opening computers so prefer some kind of testing method to see what is what.

I know I only have a couple of older versions of the SC9 so testing them is not something that I can really do myself.

Regards

[yoyo_chessboard]

tonight i have verified my 3 SC9.

first :
serial 23304332 clock: 1.49 Mhz

second
serial: 23305465 clock : 1.60 Mhz

third:
serial: 32171952 clock: 1.88Mhz

that clockrates have been measured with a logic analyser sampling at 200 Mhz. so i have a very good precision (sampling > 100 times greater than sampled signal).
they were all in factory state before i opened them.

both three have the same ROMSET
"101-1034a01" and "101-1034b02"
and the same PCB 510-1046C01

so, according to my definition of SC9A and SC9B they are all SC9A for me.
we could say :
SC9A-1.49
SC9A-1.60
SC9A-1.88

so none of mine SC9 have the PCB nor the ROMs of mike's and larry's boards.
and both 3 reply D5 to H2/H4.

and that is my problem:
on Ebay many sellers claims their SC9 are B models simply because their serial begins by 32x and that they pass some tests with better results than the basic 1.49 Mhz initial SC9A.

for me none of mine are real SC9B like mike's and larry's that i consider as real SC9B.
and on the other side we know that SC9B started in october 1983, so the first serial of real SC9B must begin with 3270xxxx at least.

i have now to do the dynamic test (colditz for example ) for my 3 clockrate.
that way we will have a solid reference:
solving time for given clock with known ROMset for SC9A boards

regards.

[Steve B]

My opinion on this.

A program has to prove that it has a different play behavior in given situations to become a version change. ie.. Super Expert A, B, C.

If any changes are as a result of speed increases then in my opinion it should be labeled as for example SC9-1 MHz, SC9-2 MHz etc. since the program itself remains unchanged.

Steve's Test would be a good indication for version changes, however even Steve's test leaves room for doubt, because the possibility still remains that the Version B because of its double or more speed increase will change its move choice because it has the ability to search the position deeper.

Therefore assuming "B" were to be twice as fast then in any fixed position tests to compare it accurately to an "A" you would also have to allow "B" half the time for these fixed position tests.

Best regards

Hi Nick
IIRC the moves were book move replies so the time makes no difference
however its not hard to test this
will give the SC9A 10 seconds to move (vs the SC9B's 5 )and see what it comers up with after 1.h4
even though the SC9B is not running twice as fast as the SC9A my guess us the SC9A will still play either 1..e5 or 1..d5

B6 Regards
Steve

Thanks, for clarifying Steve, just like you I avoid opening computers so prefer some kind of testing method to see what is what.

I know I only have a couple of older versions of the SC9 so testing them is not something that I can really do myself.

Regards

Tested both versions again
the replies to 1.h4 are NOT book moves so i recalled that incorrectly
however....
i let the SC9A think for a full minute and at no time does it ever consider 1..b6
it vacillates between 1..d5 and 1..e5

to double check .i let the SC9B also think for a full minute
it immediately chooses 1..b6 and never changes from that move

so it seems crystal clear to me

SC9B=1.h4 b6 Regards
Steve

[spacious_mind]

Tested both versions again
the replies to 1.h4 are NOT book moves so i recalled that incorrectly
however....
i let the SC9A think for a full minute and at no time does it ever consider 1..b6
it vacillates between 1..d5 and 1..e5

to double check .i let the SC9B also think for a full minute
it immediately chooses 1..b6 and never changes from that move

so it seems crystal clear to me

SC9B=1.h4 b6 Regards
Steve

Since you bought yours from a retailer as an upgraded version B. The only missing test is for other 4xxx serial number owners to do the same test to see if their computer is the same as yours.

It seems that you have found a simple test that anyone can do with their SC9's to confirm version.

Best regards

[Larry]

Yoyo asked me to put up a photo of my "non daughter board"
SC9B. I had already put my SC9's back together and didn't have time
then to take them apart again, until now. Here it is:
https://flic.kr/p/UEe5fq
In this next picture you are going to get an eyeful. The talented
Englishman Steve Braid certainly does a professional job! I don't
know if he etches his own boards or sends them off for someone else
to do them, but here is an example of his finished product: I bought this
at least partly because all the componentry and solder work is new,
unlike our 35 year old SC9's. SteveB, click on this at your own risk
Others feel apprehensive about exposing the SC9 main board, but
let me assure you, when you have dismantled a few SC9's, you can
do it blindfolded.
https://flic.kr/p/UUvZui
Larry