| ??? 03/19/10 14:41 Read: times |
#174318 - Not lucky. Errata came much too late Responding to: ???'s previous message |
The errata did not get released until maybe 6 months after several months of trial-and-error, googling, ... did find a quite costly work-around.
And the work-around was to run the processor core at less than half of full speed just to get the core to run synchronously with the USB interface and thereby not need the broken clock routing. The other alternative would have been to drop one of the two needed ethernet interfaces to feed the USB clock signal two ways into the chip - one route for the USB controller, and another route for the CPU core. All just to get the CPU core to know the exact timing of the 1ms frames. Luckilly/alas, I'm now and then working with soon-to-be-released chips. It's fun to try really new designs. But the bad thing is that not all documentation have been finalized and not all errors may have been found. |
| Topic | Author | Date |
| USB 2.0 Compliant Yet Only Full Speed 12MHz | 01/01/70 00:00 | |
| it is a common misconception... | 01/01/70 00:00 | |
| Full vs High speed | 01/01/70 00:00 | |
| USB 2.0 High Speed | 01/01/70 00:00 | |
| 480Mbps is quite fast | 01/01/70 00:00 | |
| Applications | 01/01/70 00:00 | |
| 480Mbps on the way in | 01/01/70 00:00 | |
| Maybe, maybe not. | 01/01/70 00:00 | |
| Yes, things gets complicated at higher speeds | 01/01/70 00:00 | |
| you are fortunate | 01/01/70 00:00 | |
| Not lucky. Errata came much too late | 01/01/70 00:00 | |
| What do you expect to do with data? | 01/01/70 00:00 | |
| MCU with High Speed Port | 01/01/70 00:00 | |
| other way around | 01/01/70 00:00 | |
| Remember It is the Bus Sped | 01/01/70 00:00 | |
| Still fast compared to most microcontrollers | 01/01/70 00:00 | |
| high-speed USB with 48MHz Cypress 8051 core | 01/01/70 00:00 | |
| Marketing kidding | 01/01/70 00:00 | |
re: marketing | 01/01/70 00:00 | |
| browser problem? | 01/01/70 00:00 |