Yes, but nIRQ is not 1/75. It is most like 1/150.KanedaFr wrote:which prove MCD is 1x CD driveHardWareMan wrote:SCOR period: 1/75 sec.
and you do understand why I wanted to check fi SCOR was in sync with IRQ which 1/75sec too
HDD (or flash) instead of a CD, questions...
-
- Very interested
- Posts: 750
- Joined: Sat Dec 15, 2007 7:49 am
- Location: Kazakhstan, Pavlodar
Re: HDD (or flash) instead of a CD, questions...
Re: HDD (or flash) instead of a CD, questions...
To be honest, i know the answer was on this picture...but I was unable to read it correctlyHardWareMan wrote: It is quite logical. When reader detect S1 (while S0 already in internal buffer/register) it set SCOR for one subcode byte. And reset it on first non Sx subcode. It is not necessary to set it for 2.
but still, if it read 2 frames (to get S0 & S1), it should mean SCOR should be set for 2 frames no ?
Re: HDD (or flash) instead of a CD, questions...
1/75 is what's on MCD 1 manual so it's strange.....(it's written 1 frame = 1/75s)HardWareMan wrote:Yes, but nIRQ is not 1/75. It is most like 1/150.KanedaFr wrote:which prove MCD is 1x CD driveHardWareMan wrote:SCOR period: 1/75 sec.
and you do understand why I wanted to check fi SCOR was in sync with IRQ which 1/75sec too
but it does not longer make sense since your capture
seems to prove 1/75s seems to be the MINIMUM (MAXIMUM?) value
-
- Very interested
- Posts: 750
- Joined: Sat Dec 15, 2007 7:49 am
- Location: Kazakhstan, Pavlodar
Re: HDD (or flash) instead of a CD, questions...
I understand you. But the only way to find sync is get to subcode bytes with some unique states. So to sense the sync it has to store previous subcode byte. And it detect sync and set SCOR. Then comes non sync subcode byte and it take a place of S1 subcode. And S1 take place of S0. S0 is lost. Then detector reset SCOR, since it has S1 and non sync one. Simple and workable, right?KanedaFr wrote:To be honest, i know the answer was on this picture...but I was unable to read it correctly
but still, if it read 2 frames (to get S0 & S1), it should mean SCOR should be set for 2 frames no ?
Moreover, I detected SCOR drops. But it can be because of poor CD-R quality (sometimes this game hangs...).
Yes, but I found 6.33ms periods that 2x shorter than required 13ms. But mostly they 13.3ms and longer.KanedaFr wrote:seems to prove 1/75s seems to be the MINIMUM (MAXIMUM?) value
I get it again. You was confused with this picture:
SCOR uses OR and rest of scheme uses AND. But it seems to be SCOR uses AND too.
S0 and S1 both have 1xxx xxx1, and rest subcodes don't.
Also, SCOR activation prove that detector have only one slot for subcode history, so it detect sync when S1 come in and ready to send out via subcode interface and S0 already stored in detector. Next subcode advance this FIFO and SCOR will be reset.
Re: HDD (or flash) instead of a CD, questions...
** FACEPALM **HardWareMan wrote:...
But it seems to be SCOR uses AND too.
Of course, the answer was here from the beginning :
SCOR is set when the synchronisation WORD is met.
This word is unique and nowhere else to be found.
But S0 BYTE could be found alone, as well as S1 BYTE but these 2 will never be met together...unless on the synchronisation WORD.
So when S0 is met, the CD DSP still don't know if it's S0 or a basic frame.
It sure it was S0 when S1 is met...and so set SCOR at that time.
Now, everything make sense.
- why I saw S0 as the 98th sector on some timings graph
- why SCOR is set when S1 is met
SCOR tells the host a new frame began and that the following frame will be the the first bit of P-W
Thanks HardWareMan !
-
- Very interested
- Posts: 750
- Joined: Sat Dec 15, 2007 7:49 am
- Location: Kazakhstan, Pavlodar
Re: HDD (or flash) instead of a CD, questions...
Yes. That is exactly what I trying to say, but from the viewpoint of probable realization in the hardware. In those days no one was to want to complicate any of IC part, because of many reasons, some of them are obvious.KanedaFr wrote:SCOR tells the host a new frame began and that the following frame will be the the first bit of P-W
You're welcome.KanedaFr wrote:Thanks HardWareMan !
PS
One more important thing: KS9210 tell to use falling edge of FQCK to sampling S0S1:
But MCD drive tell use rising edge of WFCK to sampling SCOR:
PPS
Now, we can use arduino for CDC controlling, CPLD for subcode decoding to track time and syncing and some I2S DAC, such TDA1543 to build true CD player, based on MCD drive. It will be interesting, heh? CPLD also can be used for multiplexing to bytes and demultiplexing to nibbles. Somebody can say that this is overkill, but it is more free way to make any test of MCD drive. What do you say?
Re: HDD (or flash) instead of a CD, questions...
Won't it be more difficult to send correct data to i2s DAC than to send serialized data from your CPLD ? (handle the data vs audio, ....)HardWareMan wrote: Now, we can use arduino for CDC controlling, CPLD for subcode decoding to track time and syncing and some I2S DAC, such TDA1543 to build true CD player, based on MCD drive. It will be interesting, heh? CPLD also can be used for multiplexing to bytes and demultiplexing to nibbles. Somebody can say that this is overkill, but it is more free way to make any test of MCD drive. What do you say?
Re: HDD (or flash) instead of a CD, questions...
Ok, I found you use the saleae software.KanedaFr wrote: What did you use to render them ?
it seems really useful to mesure time and freq as well...
Did you try the i2s plugin ?
-
- Very interested
- Posts: 750
- Joined: Sat Dec 15, 2007 7:49 am
- Location: Kazakhstan, Pavlodar
Re: HDD (or flash) instead of a CD, questions...
What do you want to get?
Re: HDD (or flash) instead of a CD, questions...
OT:HardWareMan wrote:What do you want to get?
nothing, only wanted to know if the I2S was as great as it seems (with sound save)
I discovered the saleae products this week and they seems really good...
so bad the 16channels costs 3x the 8channels version
-
- Very interested
- Posts: 750
- Joined: Sat Dec 15, 2007 7:49 am
- Location: Kazakhstan, Pavlodar
Re: HDD (or flash) instead of a CD, questions...
It is because 8x is only simple Cypress controller and it support only 24MS/s. 16x contain FPGA and can up to 100MS/s with 1 channel. So it is still cheaper than other LAs.KanedaFr wrote:so bad the 16channels costs 3x the 8channels version
I bought for $37 (more than year ago). The cheapest LA in 16ch streaming category I bought for $69.36 (april of this year). Others with no streaming or more expensive. And one more thing: Saleae and second one bundle with very powerful and handy software. Other cheaper ones don't have it and this is decisive factor.
-
- Very interested
- Posts: 750
- Joined: Sat Dec 15, 2007 7:49 am
- Location: Kazakhstan, Pavlodar
Re: HDD (or flash) instead of a CD, questions...
It's quiet down here. Too quiet.
Re: HDD (or flash) instead of a CD, questions...
I was waiting for your proto
I'm still on it but on the soft side : still working on a ROM tester.
My Mode1 ROM was working (until i break it yesterday), next step was to validate it on real hardware.
If it works as expected, I'll add the way to send specific command one at a time to check CDD control and validate the same result on CD and SD.
All my others posts on MCD forum is related to this project
I'm still on it but on the soft side : still working on a ROM tester.
My Mode1 ROM was working (until i break it yesterday), next step was to validate it on real hardware.
If it works as expected, I'll add the way to send specific command one at a time to check CDD control and validate the same result on CD and SD.
All my others posts on MCD forum is related to this project
-
- Very interested
- Posts: 750
- Joined: Sat Dec 15, 2007 7:49 am
- Location: Kazakhstan, Pavlodar
Re: HDD (or flash) instead of a CD, questions...
Like the one you described earlier with arduino, cpld and I2S DAC