To: | "" <> |
---|---|
Subject: | Re: [ts-7000] Re: Using TS-7500 SBC to collect and transfer camera data over TCP/IP |
From: | walter marvin <> |
Date: | Sun, 31 Jul 2011 09:34:21 -0700 (PDT) |
Sir, If you had read my previous posts you would see that what I propose is DMA to the Cavium, with the FPGA providing high speed timing and control. The entry point into the hardware would probably be a daughter board, replacing the processor main memory. My 30 years system software experience should make the required driver doable. My cost arguments about standalone MP projects stand Sir, I have answered your objections 3 times now. Please stop dumping on my posts thanks ]Walter From: Mike Ingle <> To: Sent: Sunday, July 31, 2011 12:04 PM Subject: Re: [ts-7000] Re: Using TS-7500 SBC to collect and transfer camera data over TCP/IP Actually, i have a lot of experience coming to group decisions about hardware. Having used the ts7300 on a past project, I found that technologics decisions regarding cpu to fpga interface left much to be desired. Have you looked at the ts7500 fpga interface? I doubt that you can move data between the cpu and fpga at 25Mbytes/s. and, there is no external buffer memory on the fpga, so only small internal buffers. What value does the ts7500 have if all you want is the data in a host computer? I say this as someone who generally prefers to work in embedded linux, and move data via ethernet. I also have been working at high data rates (65-500Msamples/s 12-14 bit ADCs) for over 10years. To get the data into an embedded controllers memory would require external hardware on the TS-7500, if the interfaces will even get the data to a DMA engine that could move the data. The
onboard FPGA at first glance appears not to have adequate connectivity, but I could be convinced I am wrong. The FPGA on the TS-7500 definitely lacks external buffer memory, which would be required if the cpu-fpga interface isn't up to the 25Mbytes/s. Data can be gotten into embedded controller memory using a "camera interfacce" found on some controllers. There are obviously a lot of ways to do it, but I have to ask why? If manipulating the data in an embedded controller is required, then by all means get the data there. If raw data at the host computer is required, then find the cheapest easiest way to get it there. Every separate development task adds complexity, So if you need to use an fpga to buffer the data anyway, then why add linux to the problem? Why? One reason might be that it is an absolute requirement that the camera data come to
the host over ethernet, then you need to find a better solution than the ts7500, which I don't think is up to the task. The beagle board xm is. Best regards Mike On Jul 31, 2011, at 8:27 AM, walter marvin wrote:
__._,_.___
Your email settings: Individual Email|Traditional
Change settings via the Web (Yahoo! ID required) Change settings via email: =Email Delivery: Digest | m("yahoogroups.com?subject","ts-7000-fullfeatured");=Change Delivery Format: Fully Featured">Switch to Fully Featured Visit Your Group | Yahoo! Groups Terms of Use | =Unsubscribe __,_._,___ |
<Prev in Thread] | Current Thread | [Next in Thread> |
---|---|---|
|
Previous by Date: | Re: [ts-7000] Re: Using TS-7500 SBC to collect and transfer camera data over TCP/IP, Mike Ingle |
---|---|
Next by Date: | Re: [ts-7000] Re: Using TS-7500 SBC to collect and transfer camera data over TCP/IP, Mike Ingle |
Previous by Thread: | Re: [ts-7000] Re: Using TS-7500 SBC to collect and transfer camera data over TCP/IP, Mike Ingle |
Next by Thread: | Re: [ts-7000] Re: Using TS-7500 SBC to collect and transfer camera data over TCP/IP, Mike Ingle |
Indexes: | [Date] [Thread] [Top] [All Lists] |
Disclaimer: Neither Andrew Taylor nor the University of NSW School of Computer and Engineering take any responsibility for the contents of this archive. It is purely a compilation of material sent by many people to the birding-aus mailing list. It has not been checked for accuracy nor its content verified in any way. If you wish to get material removed from the archive or have other queries about the archive e-mail Andrew Taylor at this address: andrewt@cse.unsw.EDU.AU