Atomic Amps is up to something...

Discussion in 'Digital & Modeling Gear' started by aleclee, Sep 29, 2016.

  1. JerEvil

    JerEvil Supporting Member

    Messages:
    4,432
    Joined:
    Jul 2, 2011
    Location:
    Las Vegas
    I tried but the most I could make show up it the "Gain" tag and what LOOKS like an "RE" before the 12. Based off the top edge, I don;t think it is a combo unless the unit is removable and the cab is like the old Atomic cabs that had holders for the POD. Also looks like same crap display as the current AF.

    [​IMG]
     
    BoogieZilla and JiveTurkey like this.
  2. JiveTurkey

    JiveTurkey Trumpets and Tants Silver Supporting Member

    Messages:
    14,288
    Joined:
    Mar 9, 2009
    Jer; please don't kill @jpage dreams.
     
    Madmax25 and JerEvil like this.
  3. mattball826

    mattball826 Member

    Messages:
    20,826
    Joined:
    Oct 23, 2009
    OK, so factor that in, plus $150 for the board and other comps, add 150 for RD/, +100 profit= 800
     
  4. StudioDevil

    StudioDevil Member

    Messages:
    128
    Joined:
    Feb 18, 2015
    I can't resist DSP talk...

    comparing a SHARC with a TigerSHARC is not as simple as comparing clock frequencies.
    Yes, the fastest TigerSHARC is 600MHz, and the fastest SHARC is 450MHz.
    BUT...

    The SHARC has hardware accelerators, and the Tiger, sadly, does not.
    What this means is that ALL the effects and cabinets need to eat up clock cycles on the Tiger, but not so on the SHARC.
    Where this is important is with impulse response cabinets, and in general FIR filters.
    The SHARC has an FIR accelerator, which can handle four (4) 1024-point IRs no problem at 48KHz...if you know what you're doing. And not everyone does, and with good reason because it's a real pita to figure out how to use them. Gladly, I've endured the process so that AmpliFire could kick some @$$.

    Now it's true that a buffered parallel process adds an additional buffer, but at 8 samples per buffer, we're talking 0.166ms (which is the same as 2 inches as sound travels). That's a small price to pay for a cost savings of over $120 per chip.

    So, say you had a 600MHz Tiger, and needed a total of 4096 points stereo IR (like 4x1024 or 2x2048), then at 48KHz you would have:

    Tiger @600MHz: 12,500 cycles per sample - 4096 cycles for IR = 8404 cycles left for the rest

    SHARC @450MHz: 9,375 cycles per sample - 0 cycles for IR = 9,375 cycles left for the rest!!!! more than Tiger!

    Of course there are FFT methods for IRs that trade latency for less cycles, and this makes for another discussion. But before you go there, the SHARC also has FFT accelerators...so, it probably trumps the tiger again.

    But, the point is, a tiger is not clearly faster than a sharc because of the sharc's hardware accelerators.
    It depends greatly on the application, and in guitar processors with cabinets as FIRs (IRs), the two chips handle about the same stuff, in reality...or at least close to it.

    In my humble opinion, they are comparable in processing power (within 10-20% of each other) in guitar modeling algos (at least mine, which are very efficient) but the price difference does not measure up to this and points to the TigerSHARCs as being a bad deal (might be why they're being dumped). And if you choose a Tiger, then you're choosing a part that will make the product a bad deal in the end.

    Anyway, what does this mean?
    Better processing for less money.
    Like AmpliFire!
     
    patriot7, mojah, crazyForce and 14 others like this.
  5. JerEvil

    JerEvil Supporting Member

    Messages:
    4,432
    Joined:
    Jul 2, 2011
    Location:
    Las Vegas
    ;)
    That's all fine and dandy BUT...

    is it a new floor unit or amp????
     
    patriot7, Greatdane, Mandrax and 5 others like this.
  6. MIJLOVER

    MIJLOVER Member

    Messages:
    2,043
    Joined:
    Mar 22, 2013
    Yeah, TigerSharc's don't really work in a florboard
    — don't they need bulky fans/cooling?

    Curious to see what it'll be! :)
     
  7. JiveTurkey

    JiveTurkey Trumpets and Tants Silver Supporting Member

    Messages:
    14,288
    Joined:
    Mar 9, 2009
    that's pretty much the only question left @StudioDevil
     
    Madmax25 and StudioDevil like this.
  8. MattAnt

    MattAnt Member

    Messages:
    1,122
    Joined:
    May 19, 2012
    Meh your wrong. Pods are ugly and harking back to something from that era makes for an ugly product...in my opinion :)
     
    bgnz likes this.
  9. MIJLOVER

    MIJLOVER Member

    Messages:
    2,043
    Joined:
    Mar 22, 2013
    This was very a interesting read! So it really is "all about the code"!

    But makes me wonder, though, — Fractal employs two TigerSharcs for the Axe Fx 2 and to Sharcs for AX8,
    And from what I've read, they Axe Fx 2 can handle a lot instances in the chain as compared to the AX8 — so is that a business decision, or it's it because everything's been build around the TigerSharcs to begin with and then by design the regular sharcks wouldn't be able to handle the same processing wise?
     
  10. chrisjnyc

    chrisjnyc Supporting Member

    Messages:
    5,015
    Joined:
    Jun 17, 2009
    I am going with bigger floor unit... maybe with 12 foot switches
     
    Mandrax and MrDavidJSmith like this.
  11. ipm

    ipm Supporting Member

    Messages:
    519
    Joined:
    Jul 2, 2012
    Location:
    Europe
    @StudioDevil
    So, you are adding more Sharcs? And you've mentioned 4 irs.
    This would really open posibilities and with vise routing, dual amp, (looper maybe) etc it would definitely push SD and AA to Fractal/Helix teritory. Hopefully, less expensive.
    Are we witnessing emergence of new big player?
     
    crazyForce likes this.
  12. StudioDevil

    StudioDevil Member

    Messages:
    128
    Joined:
    Feb 18, 2015
    another interesting point, is that even if you don't use any of the accelerators in the SHARC, it's still 75% clock speed of the TigerSHARC...they both have dual ALU/Multiplier cores, so with the right programmer, they're close. That means that if you only used 75% of the Tiger's DSP processing (or less) than you paid for it for nothing in return. But I still think they are much closer than that once you make use of the accelerators.
     
    mojah, crazyForce and iim7V7I7 like this.
  13. ipm

    ipm Supporting Member

    Messages:
    519
    Joined:
    Jul 2, 2012
    Location:
    Europe
    Very intersting recent development are IR for accoustic guitar from 3Sigma audio. With piezo they make wonders to accoustic guitar sound. This have me idea to try to use one ir in front of amp simulation (to simulate jazz box with my telecaster) and one ir for cab. Intersting results can be achieved. I'm doing this in my DAW. It would be interesting if you are making floor unit which could asssign one ir in front of everything and others wherever we choose in signal chain and to have at least dual path. Something like Helix but with StudioDevil signature.
     
    crazyForce likes this.
  14. redmonda

    redmonda Member

    Messages:
    1,140
    Joined:
    Sep 21, 2015
    @StudioDevil Thanks for all the insight (as an engineer who designs embedded systems, I always appreciate info like that). And thanks for yours (and Tom's) work on a great product. Really looking forward to whatever you guys come out with next.

    Change of topic - I know you guys owe us nothing and probably get a LOT of feature requests (I could think of one or two small ones off the top of my head ;)). Have you guys considered something like Line 6's "idea scale" where we can put in ideas for feature requests that can get upvoted to see what good ideas shake out?
     
    crazyForce likes this.
  15. jageya

    jageya Member

    Messages:
    4,254
    Joined:
    Nov 5, 2013
    wow....thanx for the info...i would love to see a new unit. I would also like to see the programming offer seemless patch switching and delay tails when going seamlessly to a new preset. Also more ambient effects like a particle verb would be cool.
     
  16. TimeSnow

    TimeSnow Member

    Messages:
    3,147
    Joined:
    Jun 1, 2015
    "RE 12"
    AMPLIFI-RE 12

    Hmmmmm
     
  17. ipm

    ipm Supporting Member

    Messages:
    519
    Joined:
    Jul 2, 2012
    Location:
    Europe
    12 Sharcs? Ha ha.
     
    StudioDevil and TimeSnow like this.
  18. mbenigni

    mbenigni Member

    Messages:
    6,567
    Joined:
    Jul 6, 2015
    Does anyone else here think of Bill Burr every time they see this thread? :D (I have the perfect meme, but I hesitate to drop an F-bomb on TGP...)
     
  19. cliffc8488

    cliffc8488 Member

    Messages:
    1,062
    Joined:
    May 13, 2006
    The TigerSHARC architecture is vastly superior to the SHARC, and it's a shame it has been discontinued. In our tests a TigerSHARC performs about 50% faster clock-for-clock. Then factor in the higher clock speed and it's about twice as fast. See here for independent benchmarks: http://www.bdti.com/MyBDTI/bdtimark/chip_float_scores.pdf . The newest SHARCs are the same as the 213xx except they have the FIR accelerator. As one can see a TigerSHARC (Axe-Fx's use the ADSP-TS201S) is about twice as fast. IIRC the current Amplifire uses two 21369 so this new product is likely using two or more 2146x as the accelerator is not available on the 2136x series.

    There are a several inaccuracies in StudioDevil's post. First of all the TigerSHARC has a more modern core and much better architecture (larger register file, wider buses, better DAG units, etc). It can do six floating point operations per clock. In practice this is two multiplies, two adds and two store/fetch instructions. This means a 4096 point IR only requires 2048 clock cycles.

    Secondly the TigerSHARC has a much wider bus than a SHARC. The TigerSHARC has a 512 bit bus whereas a SHARC has a 128 bit bus. The TS can move data around much faster and pipeline stalls due to bus contention are far less frequent. It can load or store eight words in a single clock cycle whereas a SHARC is hard pressed to do two.

    The TS also has a LOT more on-chip memory. A TS has 24 Mbits of on-chip memory compared to 5 Mbits for the best SHARC. This means more code/data in fast memory and less stalls waiting for data access. Also the TS has a superior cache unit which caches both instructions and data, as opposed to instructions only on a SHARC. There are numerous other improvements as well including better DMA engine, better interrupt handling, vastly superior 40-bit floating point support (which we use for amp modeling), etc., etc.

    Having written hundreds of thousands of lines of code for both (including coding the SHARC's FIR accelerator) my experience is that the TS is a much better chip but it is much more expensive. We moved to SHARCs for our floor processors for a variety of reasons but performance was definitely not one of them. The FIR accelerator is nice and all but it only applies to cab modeling which is a small percentage of a typical preset. On an Axe-Fx II a stereo Hi-Res cab block only uses 11% of the DSP. On an AX-8 it's still a couple percent as the accelerator doesn't handle everything needed in the block. So the net savings is less than 10%. The rest of the effects then run about twice as fast which means almost twice as many effects per preset on an Axe-Fx compared to an AX-8.

    While "all about the code" has some merit, in reality it's "all about a lot of things" including the architecture and clock speed. Algorithm complexity is also very important and it becomes diminishing returns (i.e. it takes twice as much CPU to improve the sound quality 10%). If it were just about the code everyone would still be using the original 33 MHz SHARCs. The TigerSHARC enjoyed a long run as the best DSP on the market but, alas, all good things must come to an end. The good news is that Atomic, Fractal and Line6 are all using the same DSP family in their respective floor products so it's highly unlikely Analog Devices will discontinue them and there are now more choices for the consumer.
     
    Last edited: Sep 30, 2016
  20. ipm

    ipm Supporting Member

    Messages:
    519
    Joined:
    Jul 2, 2012
    Location:
    Europe
    So, 12 TigerSharcs?:D
     
    MrDavidJSmith likes this.

Share This Page

  1. This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
    By continuing to use this site, you are consenting to our use of cookies.
    Dismiss Notice