The No.1 Website for Pro Audio
 Search This Thread  Search This Forum  Search Reviews  Search Gear Database  Search Gear for sale  Search Gearslutz Go Advanced
Lexicon 300 Reverb/Delay Processors (HW)
Old 11th December 2011
  #1
Here for the gear
 

Lexicon 300

Hi, sorry for posting a request as the first post.

But... I am looking for a Lexicon 300 v 1.2 user manual, any know where to get one?

And it also seems hard to find info on the different versions (except the v3.5 info still at Lexicon), when did they split up the DSP (for multi machines) etc.. ?

thx
Old 14th December 2011
  #2
Here for the gear
 

Ok, so now I got the Service manual (local source). But I am still looking for the v. 1.2 owners manual and info on the different versions.
Old 14th December 2011
  #3
Lives for gear
 
scott petito's Avatar
 

I may still have mine... I take a look maybe could make a copy...
As for versions ive had mine since version 1 and i still prefer that version to the split engine versions... I believe the first split was v2

Cheers
Scott
Old 14th December 2011
  #4
Here for the gear
 

That would be great and let me know if you need the SM (I have bough the M300 and 300L).

Then what would be the highest version before changing the GAL chip, or does it have to be changed with every upgrade - I mean does the v. 1.2 GAL chip only work with 1.2 EPROMs or does it work with all v. 1.x ?

once again thx.
Old 20th December 2011
  #5
Lives for gear
 

M300/L GAL Chips

Quote:
Originally Posted by the real mandak View Post
That would be great and let me know if you need the SM (I have bough the M300 and 300L).

Then what would be the highest version before changing the GAL chip, or does it have to be changed with every upgrade - I mean does the v. 1.2 GAL chip only work with 1.2 EPROMs or does it work with all v. 1.x ?

once again thx.
The Gal chip in Lex M300 version 1/1.1/2/2.2/ onwards is Rom Specific. ( It will freeze with 3.5L Rom GAL chips as these are GAL specific and many Lex units suffer from this upgrade with the owner wondering why the faceplate is frozen - mostly in 3.5L it's awaiting a LARC/Framelink if the Gal 2.2 is correct)

Basically the Version 1.02 -3/3.5 Gal Chip is the Master CPU (Allowing dual Algo's of 224XL Rich Plate + Cascade etc) and it changes the system functions in the whole set up, you need a Version 2/3 Gal chip to use with Split Algo's Software w/ Rom version 2/3/3.5/L (For Larc)
If your Gal chip is still version 1/1.1/1.2 you will only have the main Algo's (Not Spilt/Cascade/Dual Mono & others.)

The difference is huge, the 2nd series Gal chip basically adds a PCM 80 Minus Res Chords into the bargain.

Lex 300 prices are dropping lower then PCM 70's for this reason, though a Lex 300 with GAL chip 1 and Rom 1/2/2.2 etc is a very powerful unit. 480L Algo's and the Ambience / Concert Hall / Plates / Rooms etc smokes a PCM 70 into dust. A Lex M300 is worth 3 or 4 x a PCM 70 thus the Version 2/3 Gal chips with Dual Algo's are prized.

Remember all Outputs/Inputs work on Algo's so just like a 480L you can have 2 in 4 out/6 out + a Patch point which could take another unit, say a PCM 80 and use it as an extra send. W/ TC, MIDI, Automation, Format conversion, 20 Bit, 10 Sec Delay this is a beast of a unit.

Peace
P.S Hope this helps - other info also appreciated!
TLB
Old 19th January 2012
  #6
Here for the gear
 

So it would be possible to drop the v. 2.2 EPROMs into a v. 1 GAL chip and get the fixes and improvements - cool.

As far as I understand it is possible to copy the EPROMS, do any have the data needed?

THX TLB - Interesting idea with the PCM80 as insert .
Old 20th January 2012
  #7
Lives for gear
 

300 Addendum MK2

Quote:
Originally Posted by the real mandak View Post
So it would be possible to drop the v. 2.2 EPROMs into a v. 1 GAL chip and get the fixes and improvements - cool.

As far as I understand it is possible to copy the EPROMS, do any have the data needed? THX TLB - Interesting idea with the PCM80 as insert .
YES........ LEXICON REVER/FX UTOPIA YOU SPEAK........ OF CAN BE HAD!

Though..... The GAL Chip is controlling (No reference in front of me) the Z80's + etc & the Motorola Processors, so the GAL chip needs updating prior to Eprom update Chips up to 3.00/ 3.5 S/Ware - essentially (Designed to work with Framelink, I.E 1 x Larc controlling 224/480/300L all at once.) for a huge connect!

Still, if your Gal chip is prior to 2.00 you need the "GAL CHIP" upgrade Lexicon did auto to solve many 300 issues such as T/C lag & Midi freeze, Dyn Midi Preset change (List is endless - though search 224X Manuals in GOOGLE & you'll get (EVERY LEX MANUAL & SERVICE MANUAL KNOWN.) Deep Info that people try to sell on the Net & Bay. A Gal chip v.2.00 should give split algo's and "THE DIG PATCH POINT" where you can actually use the out/In of another unit and put it before the (Easy Lingo) Send/Ret you would usually use, thus making a PCM80/81/90/91 a Digital Patch Point Sending & Receiving typical FX Unit Send/Return. The Split algo GAL chip really gives you Algo's like Ponds Dither and other 3/4 of the 480L FX Algo's.....!

Are the M300 Version 2.00 Software Cascaded & Split algo's worthwhile................................................................................................................?
(The Lexicon 300M/L is 20% Faster in a Single Algorithm then 1 480L Algorithm - Hard to digest - though True!)


Hell....Yeah! as you get the added Delay, PCM80 Functions (Minus some specific PCM80 Algos specific as Res Chords), so Chambers, Halls, ambience Pitch Shifting, Dual delays Pons Dither, Compressor, Binaural Sum & Difference plus many others including LFO in the verbs and an LFO for modulating SETUP's ( Very nice indeed on the 300) and the 224XL PLATE (A GAL2./3 upgrade bonus) It is those main Dual Algo's which in a version 3 480L were packed on to 3 Boards & Carts, here, were looking at 3 boards packed with IC's on a 480L verses the 300 a Digital unit and fore runner to the PCM 80/90 Series. The M/L300 2/3 Gal chip introduces these upgrades, offering dual Mono/Split/Cascade and Main multi Algo's. Alongside Timecode, Midi, Word-clock AES /SPIDIF - And the same 18 Bit Analog In/Out as the 480L uses (ALL M/L300's Verbs are coming off the digital buss board straight to the Analog out/In Board thus the outputs are adjustable in 16 bit or 18 or 20 via a small turn wheel screw inside the unit (Though are factory set to 18 bit for the best in S/N & Aliasing scenarios (Pitch Shift howl was a problem in early M300 Units)

The 300M/L also gives you a Analog or Digital or Ab/BA Digital/Analog LR/RL "Patch Point. This is before the AD/DA's and let's you insert an FX unit or send to another console or DAT/ 1/4" tape or what ever you wish as a 2nd Send/return. Though do remember: All AES/EBU - TOSLINK -SPIDIF - ANALOG In/Out - are running! When the unit is locked be it Analog or AES, part of it's appeal for Post was it's digital operation (And even the Algo be it PONS Dither that when ...... mastering could be turning 16 Bit into 20 Bit of Head room=6dB @16 bit onto DAT. (Like Apogee UV22) Meaning you could Send a wet verb to anywhere, while doing a Mix Minus, though sending & returning Analog Snd/Rtn in a console. That alongside Random Hall, Random Ambience & 224XL Rich Plate Algo were seen as huge Post Production benefits Inc Time Code, Dynamic midi, AES-3 and Toslink (Adat Optical) and SPIDIF. All this in 1994. (3/4 of a 480L minus sampling) Please Also remember Sampling had already become affordable in low forms like the Akai S950 1000/1100/3000's so the need for sampling was negated, by tech growing Industry.

(The 480L was still used for spot sample replacement, though through a LARC and Vernier control it was a huge hassle, in time & results. Lexicon released the LFI-10 if I remember correctly to clean up the grubby DB9 AES 480L connectors, though it cost a 300M as it was - note: Ebay now has people hoping to buy these old devices for $500/1200+ since they've done some research, in fact the 480L was hopeless at sampling in storage and connecting to other devices via digital transfer, it was made for Sony PCM 501 / 1630's and the like, the the DB9 interconnects were different, and only 18 bit. AD/DA (or Less I would suspect). Dependant on boards/Rom/Ram-SME etc.

So when the M300 is born it's a unit holding 5 Algo's (And Very Brilliant they were) The AD/DA's are I understand Delta sigma Oversampling Outputs (though not Inputs) as I understand, so they hold 3 x 8 Bit Chips were arranged for 16 Bit and another 8 bits for error correction +4 bits and whoa..... we get 20 Bit resolution - thus enabling Pons Dither and Compression, + other Algo's from the 480L to be re-engineerd on to VSLI Chips (As Apple Mac's & IBM's had) though transferred into non Surface-mount boards Making the 300 Quick + efficient, adding Time Code, Midi, that could actually be locked to incoming code, and snapped to - for program or Algo changes (Great for TV/Post/Film/Automation in Music FX. V1.00 to 1.03 were those original M300 "GAL CHIPS". (Rom Chips Differed) As is usual as the M300 sold and became popular the 300L was released 2 years later. In testing M/L300 units the only way a unit could be tested was diagnostically thru an RS432/422 Port in DEB-9 format or LARC PORT (Same port as the 480L had for AES though non useable for AES on the 480L)

The shift in an M/L300 having the major Algo's of the 480L into which could hold stable TimeCode, Dynamic midi, and was "Snap To" programmable for Time Code (In All Frame rates) was huge for Post & Film. No Re-sync, Set, start, Mark print, Re-print etc. You could "Hit" the T/C points and be assured the FX would change in time with the visual pictures - something the 480L lacked (Usually with the same Algo or differing Algo preset, you'd "Hit or Snap to point & Bang! TIMECODE would lock" - End of! I do have an original Version 1.02 GAL & Lex 300 Manual "PM ME IF YOU REQUIRE IT" ( AS I loathe or certain vendors whom profit of other in this way - beware - people are selling all LEX Manuals & Schematics for all models on eBAY. Selling things on Ebay which are public domain, or that the OEM no longer supports or cares for to me is very wrong, " - GOOGLE AND YOU SHALL FIND - " It took me 3 minutes to get every Lexicon Manual alive from Lexicon Legacy!

The "BIG QUESTION" many ask is just how much is the version 2.3 GAL CHIP WORTH TO THE M/L300!

Well, I can only say that looking after around 20 + Tech notes from Lexicon on the M300, it is very worthwhile, it not only Doubles the machine's capacity in a functional way, though improves all performance aspects such as Time Code, or Midi, and even operation itself of the unit, quicker, deeper, and more playable! 3 or so of the Algo's make the 480L (Unless version 4.1 with sampling SME card and Surround LARES Cart & 224XL/Primetime 3 Carts redundant) as a Lex 480L v3.1 offers not much more in a single Algo then the 300 (Later 300 updates and other improvements enable dual machine operation, via, cascade, multi, single, etc.) Many people feel as though the 300M/L (The 300L had the GAL chip upgrade from day 1) was near born again, Delay 2/3 frame, Stereo Pitch shifting, Stereo Pitch 224XL rich Plate algo, Inverse, Chamber, Dual Delay's w/ Split, Mastering. etc New Lexicon upgraded units had a better Allocation of Analog & Digital In/Outs and a New Sticker A/B (Denoting Mach) that ALL V.2 GAL LEXICON upgraded M300's should have RE-LABELED MACH 1 /2 and A-B (Or machine 1 or 2) showing the solid split from the previous GAL chip in v1.00+ into Dual/Split/Cascade Algo's in GAL 2.0/3.0/3.5

While their are Rom Sets available for the M300 - To re-engineer the Version M300 2/3/3.5/L+ GAL chip is very much Impossible due to copy protection schemes so like recreating all of the Algo's within that particular ROM SET context without Lexicon BOM Files is impossible (GAL chips use a much more complex scheme through System interface on an Instruction for the VLSI Chip). This makes the RAM Chips for sale I.E 3.5 Rom Sets a grey-ware sale scenario & non usable for lower order 1.03 M300 GAL Chips. I have seen a few V.1.03 GAL chip M300's UNITS for sale though not working as the Sellers state they have 3.5/L EPROMS. (1.03 Gal & 3.5L Rom causes a Freeze screen on M300's as their awaiting a LARC in this instance.) Whack the old 2.0/3.0 GAL EPROM'S in and they fire up perfectly, though often V.3.5L (LARC) is thought of as the "HOLY GRAIL OF EPROMS - IT'S ACTUALLY 3.00/3.5 GAL EPROMS That make the units most stable. in my experience.

Seriously --- How many people or Super Studios have 2 x 480L and 2 x 224 and 2 x 300 on a FRAMELINK LEXICON MULTI LARC REMOTE UNIT - Near Zip! That is & was the sole purpose of EPROM GAL 3.5 for an M300 is extra Algo's... & 3.5L for LARC access. Though a 300M /L will work with a LARC in EPROM 2+ versions Software version 2.0/3.00 though it will require a 3.5L Coded GAL chip for Larc use.

The Lex 480L is getting very old indeed, and as much as our love for 224/X and 480L units go on, the 300M/L will triumph another 10 years (Mostly thanks to the 300L), as many I'm sure either have the GAL chips, somewhere stashed that never installed (I've known 3 v1.03 users get 3.0/3.5 Gal Chips in 8 months- Lexicon only ever sent these upgrades as requested, and you complained an M300 wasn't working Lex sent units to techs to be fitted to the 300M. (Even I have now a Pack of Spare Chips from Rom's to GAL to 10.3 to 3.5L ALL EPROMS) Though please, do not be fooled by an interview with X, Y or Z Engineer producer who swears on Platinum Records that a 480L or 224 was used on XY or Z's album/Single/upload. (I would never give away hard core Info on my FX Chains be they Comp/Lim/EQ/Verb/FX).

The PCM 70 saga has thrown people into a frenzy thinking Version 2.0 is "THE BEST AND ONLY PCM 70 SOFTWARE IMPLEMENTATION EVER!" PCM-70 Version 3 (Version 4 does more!) Simply updates a PCM 70 right into Dynamic Midi (To workable status) and adds Inverse as an Algo, sure...... "TILED ROOM" maybe erased though, SYS EX or hand programming can get it back in 3 minutes flat! - Seriously, around 30 commands and it's done. At the same time, a Lex 300M with GAL chip 1.03..........! They are going for $200-400 bucks more then a PCM 70 in real world pricing, I've got my stockpile of PCM's and M/L300's and I seriously suggest to any other looking to purchase and Lexicon Unit over 10 years old to AT LEAST BUY 2 UNITS, as 1 will fail in the coming time, and unlike the 224X/ the Emulation, I doubt will ever exist. If your chancing on 1 unit, be prepared for it to go bang, in a puff of Tuna smoke, H3000 Prices are skyrocketing and so are Korg Sampling Verbs & the Like, as it's "THAT SOUND THAT MADE IT HAPPEN" My only last piece of advice would be remember, someone made that unit, X,Y or Z famous, and you''ll get no points in tape op saying, "Yeah Man, we brought in the old FX unit's and viola! it was Frank (ZAPPA) or Brian (ENO) or Daniel all over again, we've discovered a new side! This stuff in the racks - Blah - Blah inset own drivel! I would look at it this way, Version 1.03 GAL M300 = $1500, Version 2/3 with Larc = $2,300 Version 2/3. M300 $1800. Those in the real know, will never sell an M300/L now as the buy price expected has just become to low. Those offloading due to the GFC and tough times in the world (Who might honestly NOT know a unit's real value) I'd be checking Inside & out.

IN My LAST WORDS LET THIS BE SIAD THE PCM-70 SAGA iS CRAZY & ON GOING - THE ANDY WALLACE PCM-42 SAGA HAS AFFECTED THE MARKET NOW FOR YEARS (AS WITH THE PM-70 being so 224XL) WELL, ALL UNITS COME FROM THE 224XL FROM THE 200, PCM42's 480L/300M/L PCM80/90/81/91/960L/92/96 etc.
Cheers
TLB.

Last edited by TheLastByte; 15th June 2012 at 09:38 AM.. Reason: Better Info
Old 21st January 2012
  #8
Here for the gear
 

TLB thx for the write-up, it really has a lot of info hard to find anywhere and it seems you are quite into the machine (and the rest of Lexicon). I have been reading some of the other posts you made about the M300, this summed a lot of it together.

But I do have a little to add about the AD/DA: The conversion is done by the BurrBrown PCM56P-K a 16bit converter; witch was the top of the pop at that time. You also find them in the TOTL CD-Players from that time (the 480L runs the PCM53).
PCM56P-K datasheet and application note, data sheet, circuit, pdf, cross reference, pinout | Datasheet Archive

This is from the service manual:




THX again, and folks- keep it coming great info.
Old 7th March 2012
  #9
Lives for gear
 

Lexicon M300 Version 3.00 Install Guide PDF File.

Interesting, though you've chosen a picture that does not show the whole story in the service manual. The Digital Board sends the Analog Inputs/Outputs and the A/D COnvertor is indeed (Early on 16 bit wide x 3) though going into the whole story briefly the digital busses to the analog outs were the filter and anti aliasing are set at either 16/18 or 20 bit dependant on dynamic range and Pitch Shift artefacts (For this reason alone there is a user adjustable wheel to move from these points, should a user find a better S/N scenario or option. The Digital Convertors are as I understand Delta Sigma Out and not on the inputs!

I've scanned this for all Lexicon M300 users.
In the mean time I really want to clear up the whole Gal Chip and Install of version 3.0 or 3.5 (Mostly into M300 with V1.00 Gal Chip) and the reason why it Freezes the Front Panel Display. That is Simply as the Gal chip & EPROM are incompatible. (Please see PDF File more V.3.0 Installation).

As stated, below you will find a PDF file regarding the Install of Version 3.0 for the Lexicon M300. It shows the Algo's added (for Cascade & Dual Mono) and also makes quite a note of how amongst the Algorithms ported over in version 3.00 is a modified version of the 224 XL Rich Plate (which has been suggested, along with the M300's Ambiance Algo, to be amongst the most pleasing for certain effects from the 224XL to the 480L, PCM70 and PCM90/91.

I have Also Scanned the Original Lexicon M300 Brochure in PDF Format (Apologies for the slightly Dog Eared nature), though it gives you a wonderful impression on the inspiration regarding the initial motives behind making a cut down 480L with 2 x Z80's and a VLSI chip instead of the 2 or 3 boards exactly like the 224/XL or 480L were jam packed and inclusive of numerous UXX Chips.

This unit really is just so misunderstood by people, even Lexicon 480L aficionado's find it hard to conceive, let alone believe just how powerful the M/L300 actually is. More so when the M/L300 using a Single Algo pops the 480L's clock by 1 & 1/2 times the power in a single instance. Couple this with 3 Inputs/Outputs, TimeCode, Midi, and the Ported effects, and I can safely say my M/L300's will be kicking on for quite some time, where as the 480L is just far to much of a risk now in Downtime, let alone the software revisions, and SME sampling boards (And Carts) all Missing, Astray or Broken!

I'm seriously fine with anyone wanting to sell (An M300) an almost Modern 480L for around $1300 USD inclusive of V.3.0 as now I have a "Store" of Spares, should 1 go down or even a GAL 2.0 chip blow. I'll quite happily swap in a version 1.00 GAL and live with the main Algorithms. With the M/L300 being supported from 1989 to 2005 I think this particular bit of Lex kit was given more life when the 300L came out straight with version 3.00 Software.

Re: the above post or my previous post. There is some confusion brought in by Nobody Special regarding the M300 Vs. the 300L, as it seems possibly that the M300 had 3 x 16 bit Chips (non Delta Sigma) and the 300L Delta/Sigma. This problem also arises as the main PDF file for repair schematics.....is..... the 300L - not the M300. (If anybody requires all the M/L300 upgrade errata & Schematic Repair Files alongside update, and known M/L300 Bug issues, please PM me.) If anybody could clarify the situation it would be appreciated! (Hello Nobody Special - Are you around?) Below you'll find the PDF Files. Peace to all M/L 300 Owners & Users! (I'm getting a message the server is busy to upload at the moment, so if I must I'll come back & edit in the PDF Files.)
Cheers,
TLB.

https://www.gearslutz.com/board/attachment.php
attachmentid=280662&stc=1&d=1331104644

https://www.gearslutz.com/board/attac...1&d=1331104644
Attached Files
Old 7th March 2012
  #10
Lives for gear
 

M/L300 20 Bit Conversion.

[QUOTE=the real mandak;7466425]TLB thx for the write-up, it really has a lot of info hard to find anywhere and it seems you are quite into the machine (and the rest of Lexicon). I have been reading some of the other posts you made about the M300, this summed a lot of it together.

But I do have a little to add about the AD/DA: The conversion is done by the BurrBrown PCM56P-K a 16bit converter - (On early units M300's this was the case it was later swapped in a service upgrade). (The M/L300 is Digital even the Analog outs run from the DSP board carrying the DIgital In/Outs.) (Yes though on the Analog Outputs); which was the top of the pop at that time. You also find them in the TOTL CD-Players from that time "the 480L runs the PCM53" (As do early M300's). Though also remember this unit was swapped as it caused Aliasing - and Pitch **** Noise in later units - the exact part is not on this computer though I will update with the service bulletin (Problem was - Lexicon only upgraded if you asked - if you told them their was a problem - otherwise all was schtung!

I understand where you are coming from, though in choosing bit rates for the 480L it runs @ 18 Bit and on the original AES/EBU DB-9 Plug (A pre-curser to the AES/EBU/Spdif spec 1985 utilizing XLR/Toslink Cables and their impedances. This originally conformed to the standards of the SONY F-1 for Mastering on UMATIC.

Moving on the the M300 (and even the L300) you will find that the actual bit rate is 20 Bit (Even on the early M300's as the A to D - D to A - Conversion process as 3 x 8 bit comparator UXX chips that were specced as I understand it to be 20 bit with the 3 x 8 rolling out 24 bits and the last 4 bits being the LSB (thus dropped for error rate protection) as opposed to the 20 MSB.

Delata-Sigma on the Outputs as I understand:

I can see & understand what you have cited in your post, though I'm not entirely sure that paragraph relates to exactly that AD/DA scenario. I will look at the 300L Repair & Diagnostic manual for clarification, (On the 300L) though I think as suggested by NBS in a post re: the 300 and using it as a convertor (After all it does have the PONS Dither and COMPRESSOR algo's that are for taking 16 Bit Paths and creating a 20 BIT sound out of the specific programs (Pons Dither & Compressor) I.E Adding 6dB of Level and using Pons Dither as a Compress/Dither - Much like UV22 Apogee style Dither -Try it!

As I've stated previously, the M300 is thought to be 3 x 8 bit convertors non Delta-Sigma Input as the Delta-Sigma conversion is on the Output This process was implemented in the PCM80/81/90/91 Series and the 300L, also it must be remembered that as the 480L was designed with DB-9 AES connection, the LFI-10 Companion AES/EBU Convertor (made much later) was made specifically for the 480L so that compliance & redundancy in Bit Length would be staved off for the 480L.

Looking through the manual of the LFI-10 it's apparent that even though bit rates could be raised, they did remain at 18 bit as the DB-9 format protocol on the 480L had to be Master! Certainly the 480L could not be easily implemented without a huge swap out of cards and that would have cost a serious fortune.

As you can see in the manual, "The Specifications of the original M300" (Not 300L) the word length & bit rate are stated as 20 Bit AD/DA. I am certainly no "Tech" though I do understand their to be a difference between the 300L and the original M300. (Possibly the 300L is all Delta Sigma) where as the M300 is a combination?

Possibly somebody will chime in (Nobody Special of course) and maybe correct any jumps led to in this thread. Though for the moment as I can understand it quickly looking at the Spec's the Digital In is Delta Sigma and the D-A is 3 x 8 Bit off the Digital Buss to Analog - I'll Post the Picture from the brochure that states this as Lexicon Puff Propaganda (They were selling their 1st AES/Toslink/Spidif Reverb after all). So perhaps we are both correct in a way!

Or very wrong?
Attached Images
File Type: png M300 DeltaSigma & 3 x 8 Bit.png (391.5 KB, 1533 views)
Old 8th March 2012
  #11
Here for the gear
 




The theoretical maximum dynamic range on a 16 bit signal is 96dB. If you look in the SM you will see that the M300 has a dynamic range of minimum 90dB, a 20 bit signal has a theoretical max of 120dB. So together with the info that the DSP board transmits two serial data lines containing 16 bit of data at 8 times the original sampling rate, I don’t think that the M300 works in 20 bit (U13 and U17 are PCM56J-K).

Now about the mystic of the AES/EBU output, it's quite simple. When you have a 16 bit signal, but need an 18 bit signal for the transmission protocol. You just add 2 bit on top of the signal and then you have an 18 bit signal, remember that the bit depth only controls the dynamic range! So by adding 2 empty bits in the top (max volume level) the only thing that happens is that the original 16 bit signal in an 18 bit environment will sound a little lower in volume - and who uses the last 2 bit anyway

But as you say, maybe Nobody Special can clear us up on the digital out.
Old 8th March 2012
  #12
Lives for gear
 

Just want chip in and say that the 300 is my favorite reverb. Even the PCM96 Surround doesn't' sound as "good" to my humble ears. (I have both)...
Old 8th March 2012
  #13
Gear Nut
 

I have a 300 with the 3.5 software, and still have the 3.0 chips, too.
Not been used for a few years since getting Bricasti (I do jazz and classical).
300 is great, but the M7 serves me better.
anyone interested in it contact me...
Old 10th March 2012
  #14
Lives for gear
 

24 bit/18/16 bit M300 through analog outputs.

As far as I could understand it the actual PCM56JK is actually the filter for the Analog outputs, and uses 2 x 16 bit as that Analog board also takes into account the PLL of the T/Code/Midi thus again making MSB & LSB more significant to the direct analog outputs/Inputs as the Lexichip were 2 x 24Bit Z80 that on the AES board are directly distributed through both parallel and Serial processing and that the Filter including the PCM56JK actually controls the signal to the Analog outputs only. Not Digital AES Board on he M300/L

The Filter is Not the AES Digital outputs, there is only 1 of those DAC filters where as there were 6 x 74Hc574's (7 actually) from the Z80's 24 bit on DSP A & DSP B which are filled directly from A-lex and B-lex and there is a mapping control (Serial/Parallel) through not directly in any way from the AES board to the NPC58813 digital filter and these 6 x 74Hc574's are all Digitally dedicated each one to a specific Digital In/Out pair (plus a spare) So from 24 Bit Z80 straight to 74HC574 (From the Schematics) is controlling the AES/Toslink/Spidif Signals not the PCM56JK (Which becomes redundant on the AES board.)

From my personal experience my M/L300 locks to the 20Bit Word length from the AES In/Out in the Aux Snds/Rtns on my Digital Console, if it were 16 bit I would be alerted as the Console is 24 bit and 16 bit signals brings in dither for encoding. Then factor that the AES signal as per AES-1985 Actually locks in this scenario. (Regardless of emphasis or Pre-emphasis.)

The other scenario apart from the AES Signal Length showing up as 20 bit and the actual Digital Filter only being on the Analog board for accuracy in Time Code & midi brings into play yet another scenario seeing as though (The 2 Boards are totally separate pieces of the 300) components have changed numerous times.

Also, apart from the 2 x Z80 & Motorola VLSI chip we also have an Analog board DAC (PCM56JK) and sitting upon that an AES board where the 6 x 74HC574's are not touched by the PCM56JK as these are AES comparative direct.

Further trouble comes when the PCM56JK is made redundant in software changes as from v.1.x onwards bugs are showing and the unit is being continually upgraded from the 256K of Ram (Buffer spills, Zipper Noise in Pitch shifting, and various other aliments means we then ramp up to Software release's; where the 2 x LexiChip Z80 boards are removed for Lexichip 2 boards.

In the progression of software enhancements Ram is introduced at 2 x 1024K and Daughter Boards are tacked on in numerous places, both with Software Rom & Ram update states (Please see documents below pertaining to the M/L300.)

So from the various release notes we see the actual PCM56JK filter is also able to be bypassed or used for 16 bit or 18 bit use again on the Analog Outputs and Inputs (This is set by the jumpers or lack their of and a trace cut.) - on the Analog Board - not the AES Board which again stays intact with 2 x Z80 Lexichip 2 directly streaming 24 bits again with MSB/LSB like the Analog though only using the complete through digital path. The LSB of the 24 Bit's is supposedly used up in various Arithmetic control routines involving the AES PLL or embedded word clock for a better example

Indeed there is a state in the Lexicon M/L300 that is a Patch point.

And this Patch Point is the difference of the Analog Board and the AES board effectively enabling a digital and or analog point (One either Before or after the AES Board or one using both the Analog summed to Left & digital to right & So on & so forth, though these are only convenience routing strategies to let the user to be chosen either before AES or ANALOG or after AES or ANALOG.
https://www.gearslutz.com/board/attac...1&d=1331378484

In General the Analog board shows the Serial data goes through the MUX into Digital Audio A or Digital Audio B with 4 Input Output Options 3 x Digital + Analog. a 2nd pair of MUX taking a feed from Digital Filter if going to the Analog Outs.

Once again the filter you speak of is rendered useless in hunkering out it's exacting design apart from V.1.x as in an upgrade notice not only is noted as used on the 480/L it is also discontinued, and an Analog devices chip of better quality is recommended. Indeed this is the case for numerous parts though the actual size of the schematics prevents viewing here, only pictures can be taken, as the scans are to large even when edited.

I can only summate that the numerous upgrade paths, and fixes recommended for the M/L300 actually make it a bit of a minefield to know what is exactly going on, to the point where the Lexichip 2 seems an unknown quantity in itself, more so on later 300L units. Indeed Parts are swapped out to such an extent that there are numerous "Don't ask Don't tell" Advisories to Lexicon or Field engineers to not upgrade should a client not request - A Rabbit hole I have seen quite a few companies get into.

Seriously, we need a Lexicon engineer to uncover this myth, though it has always been assumed, and read from numerous other posts that the unit indeed passes 20 Bit digital Audio Vs the 480L Passing 18 bits. Thus the LFI-10 & 20-20 products for 20 bit conversion are aimed at 480/L users though the whole Digital & W/Clock scenario there seems to be - more money? When will it end.

The 300 series never suffered from this scenario and the fact that the unit is based on two 24 bit Z80 Lexichip/2's and (7) 6 Direct to digital on the AES Board convertor 74HC574 IC's makes me think it is a 20 bit unit in digital utilizing the AES board and in V.3/3.5 even 18 bit on the Analog IN/OUT as far as I understand the material at hand upgrading and implementing at least 8 fixes with major Software updates, hardware updates, inclusive of the fact that in version 3.5 with the Spilt/Cascade Algos' we have Pons Dither, Stereo Adjust, and Compressor.

Should anyone else be able shed hard light on the AES Board or rather the Analog board and it's filter It would be welcomed!

P.S Please also not that the Lexichip's are updated to Lexichip 2 and then updated yet again to Lexichip 2A's

Peace
TLB.
Geez that was exhausting reading! Phew!

https://www.gearslutz.com/board/attac...1&d=1331378780

https://www.gearslutz.com/board/attac...1&d=1331378780
Attached Images
Attached Files
File Type: pdf m300-300L_repair_procedure copy.pdf (321.1 KB, 1244 views)
Old 14th March 2012
  #15
Lives for gear
 

Lexicon M300/L 20 Bit Lexicon Knowledge-base

Hey TRM just thought this would close the door on the M300 being 16 Bit:

http://www.lexiconpro.com/knowledgebase.php?product=119&topic=&entry=1035
Direct Quote from the Lexicon Knowledge Base Q & A asking numerous questions and of those quite a few centre on the M300's Bit Status also including the M300 Strips all DAT ID's due to the word length. Please see the link above to Lexicon Pro knowledge base.

M300 :

Question:

What happens if you send a 24 bit AES/EBU word into the M300?

Answer:

The M300 will process AES/EBU signals of up to 20 bits in length. A 24 bit word sent to the M300 will be truncated down to 20 bits. For optimum audio performance it is suggested that the 24 bit signal be dithered down to 20 bits prior to sending it the M300.


It has been assumed since the unit came out it was a 20 bit capable machine, this is a question from the old (Ancient Lex Knowledge-base) and it states clearly, 20 bits. I think your point about the analog board DAC is indeed true, though the fact you can remove, (or Trace/Cut) and then switch from 16 Bits to 18 Bits to No bit length on the Analog board does seems strange; the 90dB scenario another, I just see no published Digital AES Board Specs anywhere save Errata & service Schematics save the control Digital Convertors for that AES board itself & the through put of MSB/LSB (Most significant byte to Least) chewing off 2 Bit's for correction & 2 Bits for Word Clock embedded in the M300/L.

The last word really has to be that the M/300/L seems as swapped out in parts be it Rom/RAM Hardware boards, Daughter Boards, RAM Boards, LexiChip 2 A Boards and the Software as is the 480/L though the number of IC's in the 300 equate to the 50 used in a "Hard Bypass" on the 480/L.

Peace.
TLB.
Old 6th October 2016
  #16
Lives for gear
 

Addendum:

I will also add here that Lexicon in all its products never really made full use of the full dynamic range nor Sonic depth and ability, this is for many reasons.
1. Lexicon Algorithims such as the PCM -70 were sloped off at 15Khz, the H3000 using a near same PCM chip and Muratta filter was 20Khz roll off.
2. Rich Plate was never designed for such brightness
3. Only the Aliasing effects of the Pitch Shift program seem to show an upper signal strength though again nowhere near a DSP4000 at 20/24 Bit pitch shifting
4. The majority of Algos simply sounded far too shrill and or the added and extra high frequency content ruined the Algos in higher order SNR and THD.
5. The M/L 300 as a unit solo requires many digital functions to happen simultaneously thus meaning a buffered start and self test routine are near bypassed.
6. As these "Normal Diagnostic Routines" are bypassed the 300 enters a simple on and off routine minus bit and Dram or EPROM or Memory checks. This firmly accounts for the M/L diagnostic software availBle via the 9 pin D-Sub reserved for LARC control.
7. As the unit was released in 1989 and the L300 in 1993 only then do we see just how many additional daughter boards and upgrades were given to even then field purchased M300 units including LARC and Split/Cascade mode on the 300 unit.
8. Lastly, always register a product even S/H purchased it helps the OEM keep you up to date with upgrades and eliminates advertising said upgrades that you might not or will not notice etc. The most important being Software and Firmware upgrades!
9. Before buying an older M300 check the date code and see what Daughter board upgrades the unit has. As version 3.5 software will run on a version 1.02 GAL chip though very slowly and very, very much compromised as the O.S is unable to instruct the unit fully. Date code 11/12/92 means 12th day of November 1992. (Example only) no date code? Pop the lid, look inside!

TheLastByte.
2016
Old 6th October 2016
  #17
Lives for gear
 

Quote:
Originally Posted by the real mandak View Post



The theoretical maximum dynamic range on a 16 bit signal is 96dB. If you look in the SM you will see that the M300 has a dynamic range of minimum 90dB, a 20 bit signal has a theoretical max of 120dB. So together with the info that the DSP board transmits two serial data lines containing 16 bit of data at 8 times the original sampling rate, I don’t think that the M300 works in 20 bit (U13 and U17 are PCM56J-K).

Now about the mystic of the AES/EBU output, it's quite simple. When you have a 16 bit signal, but need an 18 bit signal for the transmission protocol. You just add 2 bit on top of the signal and then you have an 18 bit signal, remember that the bit depth only controls the dynamic range! So by adding 2 empty bits in the top (max volume level) the only thing that happens is that the original 16 bit signal in an 18 bit environment will sound a little lower in volume - and who uses the last 2 bit anyway

But as you say, maybe Nobody Special can clear us up on the digital out.

The RS422 or AES section requires those 2 bits to almost parallel input and output
1.AES
2.S/PDIF
3.TOSLINK
4. The SMPTE comes off the Analogue Board which again takes its signal from another PCM56/J/K/M Chip hence some saying their preference is AES out or Analog. Also don't forget we are not dealing with a 16 bit signal though a 24 bit Delta thus we are taking away bit depth Not adding it. This is why the hard RS422-AES equates to 20 Bits Delta Sigma X 3 chips at 8 bit each. We must also remember we are talking Reverberaton and FX so in an FDN network full of all pass and nested filters the least noisy unit wins thus the best SNR not the most capable makes the 300 Reverb and Effects unit sound best. It's not an Apogee Convertor striving for excellent "Conversion" just decent conversion hence a lower dynamic range.

I hope that helps!

TLB
Old 17th October 2016
  #18
Lives for gear
 
Oldone's Avatar
It's been awhile since this thread began but TheLastByte, thank you for all this information and links. The prices continue to fall and I just picked up a GAL v2.01 with the 3.5 chipset for $899. There were two at the store V1.1 and V2.01. I figured I would go with the latter as they were the same price. I have several project studio reverbs and this one, well it immediately out classes my TC Electronics M3000 which has been my go to box. For lack of a better description it sounds like a pro recording when I put vocals through the box with minimal tweaking required. Another added bonus is that it came out of a TV production house here in Los Angeles and there were a lot of bonus custom verbs in the User Presets section. Pinky and the Brain, Deep Space 9, Star Trek Generations and some other well know shows have custom verbs stored that I can now use.

Now I need to find me a LARC. I was able to create a midname file in Protools to quickly access all the presets but I would like to easily access the parameters. If anyone has one or knows where to get one, let me know.
Old 18th October 2016
  #19
Lives for gear
 

Quote:
Originally Posted by Oldone View Post
It's been awhile since this thread began but TheLastByte, thank you for all this information and links. The prices continue to fall and I just picked up a GAL v2.01 with the 3.5 chipset for $899. There were two at the store V1.1 and V2.01. I figured I would go with the latter as they were the same price. I have several project studio reverbs and this one, well it immediately out classes my TC Electronics M3000 which has been my go to box. For lack of a better description it sounds like a pro recording when I put vocals through the box with minimal tweaking required. Another added bonus is that it came out of a TV production house here in Los Angeles and there were a lot of bonus custom verbs in the User Presets section. Pinky and the Brain, Deep Space 9, Star Trek Generations and some other well know shows have custom verbs stored that I can now use.

Now I need to find me a LARC. I was able to create a midname file in Protools to quickly access all the presets but I would like to easily access the parameters. If anyone has one or knows where to get one, let me know.
Hi there, that's great news, as for midi,ni can suggest sound diver or Midi Quest though you'll need to run it at 32 bit and on OS X snow leapard apart from that the 300 should come close to system ex of the PCM 91. If not you can scope out that program and create one on Midi Ox on iOS or OS X etc.

Good luck!
TLB
Old 17th November 2016
  #20
Gear Maniac
 

Hey OldOne-
Three quick questions for you. I'm looking to potentially add a 300 with 3.5 and 2.01 OS…
a.) Do you notice any speed/UX issues with the older 2.01 OS?
b.) Assuming that combo has the Rich Plate?
c.) Can you confirm performance digitally through AES at 24 bit? Wondering how this may sound comparatively as I would have to run through the AES on my Prism.

Thanks for any help in advance
Old 17th November 2016
  #21
Lives for gear
 
Oldone's Avatar
Quote:
Originally Posted by bink View Post
Hey OldOne-
Three quick questions for you. I'm looking to potentially add a 300 with 3.5 and 2.01 OS…
a.) Do you notice any speed/UX issues with the older 2.01 OS?
b.) Assuming that combo has the Rich Plate?
c.) Can you confirm performance digitally through AES at 24 bit? Wondering how this may sound comparatively as I would have to run through the AES on my Prism.

Thanks for any help in advance
a. No speed issues at all. Very snappy in fact. No UX issues either.
b. Rich Plate is there
c. I am running analog off the patch bay so can't comment on the AES.

If you are in LA...the Pasadena GC had one left for $895 in the back of the store. Probably will be gone 15 minutes after I post this.
Old 18th November 2016
  #22
Gear Maniac
 

Thanks for the input. I'll give em a ring and check that out.

Anyone else using there's via AES in a 24 bit fashion?
Post Reply

Welcome to the Gearslutz Pro Audio Community!

Registration benefits include:
  • The ability to reply to and create new discussions
  • Access to members-only giveaways & competitions
  • Interact with VIP industry experts in our guest Q&As
  • Access to members-only sub forum discussions
  • Access to members-only Chat Room
  • Get INSTANT ACCESS to the world's best private pro audio Classifieds for only USD $20/year
  • Promote your eBay auctions and Reverb.com listings for free
  • Remove this message!
You need an account to post a reply. Create a username and password below and an account will be created and your post entered.


 
 
Slide to join now Processing…
Thread Tools
Search this Thread
Search this Thread:

Advanced Search
Similar Threads
Thread
Thread Starter / Forum
Replies
Mike H / High End
70
MFPhouse / Electronic Music Instruments and Electronic Music Production
7
Musicartstudio / Gearslutz Secondhand Gear Classifieds
0
rr1073 / High End
4
Mike H / Gearslutz Secondhand Gear Classifieds
1

Forum Jump
Forum Jump