Just How Fast Can a P1 Go?

Comments

• kwinn Posts: 8,223

  2019-02-16 - 03:50:49

  AwesomeCronk wrote: »

  Seems to typically run at 80Mhz, but I’ve heard of it run at up to 96Mhz. Is that the limit? How might this adversely affect the circuitry inside?

  
  AFAIK P1's can and have run at 96 and 100MHz reliably. I believe some people have had them running at even higher speeds, but I do not recall what speed or who did so.

  In science there is no authority. There is only experiment.
  Life is unpredictable. Eat dessert first.
• Genetix Posts: 1,397

  2019-02-16 - 04:35:17 edited 2019-02-16 - 04:36:27

  AwesomeCronk,
  
  Parallax sells this faster crystal if you want to replace the usually 5 MHz found on most boards.
  https://www.parallax.com/product/251-06250
  
  5 MHz x 16 PLL = 80 MHz
  6.25 MHz x 16 PLL = 100 MHz
  
  You should be posting these Propeller related questions in the Propeller 1 forum.
• Genetix Posts: 1,397

  2019-02-16 - 05:02:31 edited 2019-02-16 - 05:05:35

  AwesomeCronk,
  
  Here you go.
  https://forums.parallax.com/discussion/117476/maximum-clock-speed-for-the-propeller
  http://forums.parallax.com/discussion/149239/maximum-reliable-clock-frequency
  https://forums.parallax.com/discussion/143703/propeller-clock-speed
  https://forums.parallax.com/discussion/102260/maximum-input-clock
  https://forums.parallax.com/discussion/131706/114mhz-propeller-1-0-55/p1
  https://forums.parallax.com/discussion/122680/7-3728mhz-crystal-wont-run-at-16x-but-would-a-faster-one-run-at-8x
  https://forums.parallax.com/discussion/117864/the-clock-clock-multiplier-and-crystals
  ... Chip chimes in on this post about the PLL for those who are interested.
  https://forums.parallax.com/discussion/113808/the-6-250mhz-crystals-are-in-run-your-propeller-at-100mhz/p1
• Mike Green Posts: 22,891

  2019-02-16 - 05:07:59

  The P1 was originally spec'd for 80MHz over the full military temperature range (-55C to 125C). It has been used in products at 96MHz and also used at 100MHz. Page 32 of the Propeller Datasheet gives a graph of the highest operating frequency vs. temperature. Details of the test procedure and conditions are given there.
• Cluso99 Posts: 14,900

  2019-02-16 - 06:27:20

  I nearly always use 104MHz on my boards. I have specifically designed them for that speed.

  My Prop boards: P8XBlade2, RamBlade, CpuBlade, TriBlade
  Prop OS (also see Sphinx, PropDos, PropCmd, Spinix)
  Website: www.clusos.com
  Prop Tools (Index) , Emulators (Index) , ZiCog (Z80)
• AwesomeCronk Posts: 586

  2019-02-16 - 14:28:53

  104Mhz??!! Good grief! I’ve never gone beyond 80Mhz!

  That's not an explosion! That's an unscheduled rapid disassembly!
• Cluso99 Posts: 14,900

  2019-02-17 - 03:06:36 edited 2019-02-17 - 03:07:23

  Sapieha was the master. He had my TriBlade running reliably for months? at 115MHz.
  For some unknown reason, DIPs run a little faster than QFPs.
  
  I have never found a problem at 108MHz but I never continued as 104MHz was fine.

  My Prop boards: P8XBlade2, RamBlade, CpuBlade, TriBlade
  Prop OS (also see Sphinx, PropDos, PropCmd, Spinix)
  Website: www.clusos.com
  Prop Tools (Index) , Emulators (Index) , ZiCog (Z80)
• AwesomeCronk Posts: 586

  2019-02-17 - 13:05:07

  Cluso99 wrote: »

  For some unknown reason, DIPs run a little faster than QFPs.

  
  Maybe the pins being larger allows the power supply to be more reliable?

  That's not an explosion! That's an unscheduled rapid disassembly!
• jmg Posts: 13,451

  2019-02-17 - 19:37:56

  AwesomeCronk wrote: »

  Cluso99 wrote: »

  For some unknown reason, DIPs run a little faster than QFPs.

  
  Maybe the pins being larger allows the power supply to be more reliable?

  
  I'd expect it to have more to do with thermal resistance. The DIP is a much larger radiating surface, than a QFP.
  P2 seems to benefit from the thermal pad and 4 layer PCBs - did anyone test the P1 QFN (now NRND) package on 4 layers for speed ?
• altosack Posts: 130

  2019-02-19 - 14:33:43

  I've used 7.3728 MHz crystals for ~118 MHz, which worked fine because all my cogs were in WAITCNT loops of one kind or another, and it only got slightly warm to the touch. At the time, 7.3728 MHz crystals were cheaper than 6.25 MHz crystals, which is what I usually use now. Unfortunately, I don't use enough of them to worry about the cost that much; maybe that will change with the P2 !
  
  That was on project boards (TQFP).
• AwesomeCronk Posts: 586

  2019-02-19 - 21:48:54 edited 2019-02-19 - 21:49:26

  (Whistles in appreciation)
  There are some crazy speeds to be attained by the p1!

  That's not an explosion! That's an unscheduled rapid disassembly!
• Mark_T Posts: 1,975

  2019-02-28 - 18:17:45

  98.304MHz works reliably for me with I2S applications - thats 16x a 6.144MHz crystal.