CPU load on MKS SBase(AT2560) with Linear advance, bed skew, junction dev and s-curve running?
Posted by dasflux
|
CPU load on MKS SBase(AT2560) with Linear advance, bed skew, junction dev and s-curve running? May 22, 2018 11:39PM |
Registered: 3 years ago Posts: 14 |
Some guys in the Tarantula community were wondering about the processor limits we can expect running some of the latest marlin features together such as linear advance, s-curve, junction deviation, and bed skew.
The Tarantula uses a MKS Sbase 1.2-1.4. My first question is if the Atmega 2560 is 8 bit. What makes this board 32bit? The delta little monster uses a MKS Gen and it does pretty well but a lot of people upgrade to the duet so im assuming the obvious here...
Our question is basically the CPU load we are going to get if we turn on all these features, if we even can. Can anyone give me some guidance? It would be cool if we could actually get a real time CPU read out in marlin and see(anyone know of such a thing?)
The Tarantula uses a MKS Sbase 1.2-1.4. My first question is if the Atmega 2560 is 8 bit. What makes this board 32bit? The delta little monster uses a MKS Gen and it does pretty well but a lot of people upgrade to the duet so im assuming the obvious here...
Our question is basically the CPU load we are going to get if we turn on all these features, if we even can. Can anyone give me some guidance? It would be cool if we could actually get a real time CPU read out in marlin and see(anyone know of such a thing?)
|
Re: CPU load on MKS SBase(AT2560) with Linear advance, bed skew, junction dev and s-curve running? May 23, 2018 02:31AM |
Registered: 8 years ago Posts: 4,678 |
MKS Sbase 1.2-1.4. is a smoothieboard clone
It runs a 32-bit Cortex-M3 LPC1769 CPU (which apparently can run the new marlin 2.0)
This is what makes it 32 bit.
Your probably confused with MKS naming conventions as they are not very logical..
MKS Sbase vs MKS Base, the latter which uses 8 bit Atmega 2560
There is no OS as such on these controllers, you cant get a real time CPU read out... All you can do is profile the code and see how long it takes to run.
It runs a 32-bit Cortex-M3 LPC1769 CPU (which apparently can run the new marlin 2.0)
This is what makes it 32 bit.
Your probably confused with MKS naming conventions as they are not very logical..
MKS Sbase vs MKS Base, the latter which uses 8 bit Atmega 2560
There is no OS as such on these controllers, you cant get a real time CPU read out... All you can do is profile the code and see how long it takes to run.
|
Re: CPU load on MKS SBase(AT2560) with Linear advance, bed skew, junction dev and s-curve running? May 23, 2018 03:22AM |
Registered: 4 years ago Posts: 5,232 |
|
Re: CPU load on MKS SBase(AT2560) with Linear advance, bed skew, junction dev and s-curve running? May 23, 2018 05:53AM |
Registered: 5 years ago Posts: 14,152 |
Quote
o_lampe
My best guess is, the sd-card readout is the bottleneck, not the CPU. The Duet uses the hardware port for SD-card, not sure about Smoothie or it's clones...
The Smoothieboard and clones use SPI access to the SD card, same as the 8-bit boards. Only the Duets use fast 4-bit parallel access to the SD card. However, even with SPI access the speed of reading the SD card during printing shouldn't be a bottleneck. The main benefit of fast SD card access is that it allows high speed file uploads to the SD card over the network.
BTW there is also an experimental port of RepRapFirmware for LPC1769 boards, see [forums.reprap.org].
Large delta printer [miscsolutions.wordpress.com], Robotdigg SCARA printer, Crane Quad and Ormerod
Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].
Sorry, only registered users may post in this forum.