Should I be disappointed with SSL?(alphalink AX)

[Svart]

we should explore cable drivers as A/D input buffers.

I think you and I see this in the same light. I also frequently suggest giving the new ADSL/video opamps a try in audio. They may have 1000+v/us slew and all that fancy stuff that folks denounce, but they are very accurate, they don't ring or overshoot much, their settling times are outstanding and they are generally designed to drive cables and capacitive circuits directly.

Their prices are coming down due to the increased usage and we can only expect cheaper/better parts in the future, but still are 20-100x the price of the jrc4565.

In the future I might try one of my favorite opamps in place of the JRC but I need to try the system first.

I should be able to report back something today.

 

[Svart]

The opamp of the minute is the lm617X series. I also love the AD825 and other CFB AD8XX models too. The THS series from TI is also very good, I think D. Self also mentioned one of these once.

EDIT: the lm617x will drive 75R and 50R with ease..

I've also found that you can get them cheaper than the opa series sometimes too..

 

[Snatchman]

SOT....Question: I don't have a lot of experience with DIY only reading and enjoying the stuff you guys do! I've pondered this and since this about the SSL has come up,.." Which is better.." A poorly designed circuit using premium components or a good designed circuit using inferior components?"... .. Just ONE inquiring mind that want to know..Thanks!

 

[Svart]

Personally I think you can have both or a mixture of either. Someone can always screw up a circuit that uses the "best" parts, but they can also make something truly great. The OPA parts are usually noted as being good, but I'm not a fan of them so much, however that is beside the point here. SSL uses the 5534/5532 a great deal and well, you already know the results, they speak for themselves. They also use the OPA FET input opamps, but only where they need a FET based opamp and don't want to use the TLOXX series opamps(even though some of their newer stuff seems to use these opamps in the audio chain...).

The OPA series don't like to drive too much load and can sound pretty awful when stressed, just one of the reasons why expensive parts might not sound good and why the blind swapping of parts might not yield good results.

I tried a lot of opamps in my console and for the most part have gone back to the Signetics 5532 that were in the console in the first place. they do have a different sound but they definately sound the best IMHO. I did leave a couple channels as opa2134 without coupling caps because of their low end. It has some kind of gritty low with a scooped mid that sounds great on bass, but again this just happened to work good for this sound and not for much else.

 

[Rochey]

ADSL Line drivers as Opamps???



for such an application, you might like the look of the TPA6120A2 or the OPA1632... both have roots in the ADSL/Video world.

/R

 

[Svart]

I have a few 1632 laying around. I'll try to get a couple breadboarded. I am still seriously thinking about racking the tascam analog stuff with some ADAT chips just to see how they sound outside of the unit.

So far I've done some simple listening tests with the SSL box. It took forever to figure out the "mixer" software and get Reaper to see the audio streams then get usable templates built.

The unit sounds fine. Pretty neutral, nothing really special so far, a bit more clear than the tascam.

more later

 

[Rochey]

[quote author="mediatechnology"]Rochey: Had a quick look at the OPA1632 and it's pretty impressive. Only $1.75 at 1K and available. Just what the Dr. ordered for A/D driving. I'd be curious to see if svart could do some measurement or listening tests between a 4565 and the 1632.[/quote]

I don't want to thread hijack - I'll just add, that OPA1632 is an awsome little chip - it's used on all our AD AVM's now, as it's performance exceeds what our ADC's can do (even the PCM4222).

/R

 

[Svart]

well my echo mia soundcard crapped out last night. the inputs distort and oscillate randomly now. This seemed to be related to the lightning we had here...

anyway, I was forced to buy an M-audio card the 192 something or other.. It has an akm 5385 adc, 5532s on the inputs and cheap caps.

anyways, the mia still outputs and the digital input still works. I think i'll whip up an A/D with spdif out using the 1632 and piping it right into the card. that should tell us something at least..

I need to do a lot more testing with the SSL stuff before I can really talk myself into doing any modding.

:thumb:

 

[Svart]

Ok another thing about the SSL, the damn thing is picky about terminations. if the cable is even slightly loose or noisy, you WILL hear it. it will easily pick up RF noise.

I guess that's a good thing, at least we know the bandwidth is there.

With proper connections, the unit is dead quiet, somewhere around -80 to -90db.

I like it so far. it's a little more open than I'm used to soundwise so I am still getting used to it.

the thing I don't like so far are the stock EQ plugins. they don't sound all that great. The console EQs sound much better to my ears. On a loud rock song I bet you can't hear it though.

I needed to hook the tascam back up finish up some work and noticed that it sounded much more "rounded" than the SSL.

 

[Svart]

So now I have to send the AX back to be "upgraded" with newer firmware, seems that there was some kind of bug in the firmware that won't let you use 88.2k or 96k sampling rates, and it's true.

waiting to hear back from SSL and my sales guy on this..

 

[Svart]

It's in circuit programmable logic, looks like FPGA or CPLD, so the firmware would really be HDL. I have a programmer for this but I don't think SSL would send me the raw code to load.. :green:

 

[Svart]

when the unit gets back I have some 6172s to install as well as going to trace the circuit and get a good look at the clock circuit to see how good it is.

 

[Svart]

their literature says "possible problems using 88.2/96 with mixpander.." Maybe their prototype testing units worked in their testing configuration but later found that other configurations don't work?

Maybe a corrupted file got loaded into the programmer?

Who knows, but they are paying for the shipping and the upgrade is free.

I still need to call them about the turnaround time though, I'll ask them about the problem when I call.

 

[Svart]

It's going to SSL New York. apparently they have an office/repair shop there. According to my dealer, the only other unit in stock was also from the same batch that shipped with firmware problems and did not have the update either. They also said that since I registered the unit already that it cannot be exchanged, citing policy and such.

I'm sure that the firmware update is "in circuit" via JTAG, at least I was pretty sure I saw the header for it when I took a look inside. If it's FPGA based then they are programming an EEPROM/flash, if it's CPLD then they are programming the HDL directly into the CPLD, if it's DSP or MCU then they are likely loading into flash.

However I do think I remember seeing a socketed PLCC or something like that.. maybe I might be lucky and get them to send me a new IC instead.

I'm going to ask them all about it, we'll see if they answer my questions.

Also, according to SSL they updated the CODECs to "better" ones, it's possible that AKM claims 100% compatability between those and the older CODECs but in fact might need a bit flipped or something to be totally compatable. AKM is good at making mistakes and typos in their datasheets from what I've worked with.

 

[Rochey]

*cough* - really?

 

[mikep]

do you have any solid info yet on how the thing actually SOUNDS?

 

[Svart]

Sure do. It sounds fine. it's exactly as described, not hyped or boomy, pretty flat across the audio band. I still need to do side by side comparisons but I haven't heard anything offensive yet.

I'll try to get some clips up soon as well.

 

[Svart]

I talked with SSL for a few minutes. They were happy to tell me that it's a firmware flash AND a board swap, but it seemed that they didn't want to tell me what board they were swapping.

 

[Svart]

who knows, maybe they needed to replace the digi board and THEN flash it, but it was said that they were going to "flash a board and change another board".

Since there are only two boards, A/DD/A and digital/interface, that would mean that they are changing the analog boards, maybe to a newer/different codec? I don't know but they said they would have it back to me within 3 days.

We'll know more then.

 

[SonsOfThunder]

Svart,

Take lotsa pics before you send it. Then you will know what they changed for certain.

HTH!