I'm excited about this for an embarassing reason. I have never even powered up the Spinneret that I bought ages ago because I've never gotten around to making a power cable for it. This one I can just plug into a USB port and program away!
Have you decided on a size / part number for the QuadSPI flash device yet ?
That's an interesting question. RossH seems to have discovered that the 4k erase command doesn't work correctly in QPI mode on the Winbond chip that is on the Propeller Memory Card. I have tried using it as well with similar bad results. I've send a question to Winbond to see if they can tell me what I'm doing wrong but haven't heard back yet.
That's an interesting question. RossH seems to have discovered that the 4k erase command doesn't work correctly in QPI mode on the Winbond chip that is on the Propeller Memory Card. I have tried using it as well with similar bad results. I've send a question to Winbond to see if they can tell me what I'm doing wrong but haven't heard back yet.
Does the command work in std spi mode ?
Maybe they limited the erase commands to be via Std SPI only, to protect them a little from the efforts to streamline QSPI ?
Does the command work in std spi mode ?
Maybe they limited the erase commands to be via Std SPI only, to protect them a little from the efforts to streamline QSPI ?
Yes, it works in SPI mode and also in Quad SPI mode. The datasheet claims it also works in QPI mode but RossH found that it took about 30 seconds to erase a 4k block and I had similar results.
Yes, it works in SPI mode and also in Quad SPI mode. The datasheet claims it also works in QPI mode but RossH found that it took about 30 seconds to erase a 4k block and I had similar results.
Oh, so it is a command issue, more than a QuadSPI one ? It is 30 seconds, in either mode ?
The Winbond data I have 25Q128 mentions 25 sec typ as a chip erase time - does it erase everything ?.
Oh, so it is a command issue, more than a QuadSPI one ? It is 30 seconds, in either mode ?
The Winbond data I have 25Q128 mentions 25 sec typ as a chip erase time - does it erase everything ?.
I'm not sure if it erases the entire chip when you ask it to erase 4k. If so, it could take up to 50 seconds. The 4k block erase should take a maximum of 400ms.
As I'm working through Season 3 of Babylon 5, I'd guess you've found a way to use a Propeller and this add-on to replace human telepaths as the operating system for Shadow attack ships. Way to compromise Earth Force security, dude.
B5 season 3 hey? But which pass? I think I've watched all 5 season about 3 times. BTW, Propellers already play havoc with the Vorlon's sense of order here on Earth. YESSSS. If you go to Prop'doom you will not return.
BTW.2 - OT - Are you still using the W5100? After sorting out a few things with the W5200 I'm much happier with this chip.
Personally, I would much rather see that 40-pin, 16-bit connector broken up into two 8-bit connectors without that big, unusable gap in the middle, each with its own power and ground connections. I'm not a huge fan of the QS pin layout. It makes prototyping difficult, since there's only one set of power/ground pins way at the end, which limits one to a single multi-function add-on board, rather than a couple single-function boards. The latter is the more modular approach to prototyping. And, if you need all 16 pins, a double-size daughterboard would do the trick. It's all about providing options, IMO.
Of course, my personal preference would be two 2mm daughterboard connectors like the one on the original Spinneret. That would leverage Parallax's investment in that format and provide plug-in compatibility with the boards that already use it.
Personally, I would much rather see that 40-pin, 16-bit connector broken up into two 8-bit connectors without that big, unusable gap in the middle, each with its own power and ground connections. I'm not a huge fan of the QS pin layout. It makes prototyping difficult, since there's only one set of power/ground pins way at the end, which limits one to a single multi-function add-on board, rather than a couple single-function boards.
True, but I suspect the QS pinout is rather locked-in now.
What about a Raspberry pi connector in a corner ?
If there is not room for a Raspberry Pi 26 pin P1, then maybe a 10 pin P5 ?
That has i2c or i2s options, and i2s on the Pi looks to have quite good HW support. (duplex FIFO depth, at least)
I'm not sure of the drivers, or if i2s can be used as pure digital IO transport (a RPi expert may know ?) , but the stated 3.072MHz looks (just?) do-able on a Prop ?
3MBd duplex (or even half duplex) is quite a decent slave link, all on 10 pins .
David,
Looks like what I predicted and what I was hoping for but didn't say (the flash memory and sd card slot). I had assumed a bunch of the pins would be used already, but I figured some QS addons might work if they didn't overlap. The QS connector could be reversed on this board so that QS addons hang off the edge instead of overlapping... At least that way someone could make an addon board that would work on either this or the QS (assuming they only used P0-15).
Anyway, really happy it has the W5500. Looking forward to getting one when they are available!
I made a W5500 driver for Parallax and open it on my github. https://github.com/jbkim/Parallax_W5500
It’s based on Mike’s code but I don’t use the Asm because of better understanding about the SPI operation.
I'm not a huge fan of the QS pin layout. It makes prototyping difficult, since there's only one set of power/ground pins way at the end, which limits one to a single multi-function add-on board, rather than a couple single-function boards.
I'm glad I'm not the only person with this opinion. I much prefer the Propeller Platform form factor over the Quick Start pin layout.
I made a W5500 driver for Parallax and open it on my github. https://github.com/jbkim/Parallax_W5500
Its based on Mikes code but I dont use the Asm because of better understanding about the SPI operation.
-JB
Hi JB, welcome to the Prop forum although it seems you are not new to the Propeller obviously. Been lurking hey? Well done with writing the drivers but I couldn't see where it is serving up any web page files?
How did you find the W5500 chip to work with? If you have worked on the W5200 then how does it compare? I'm looking at using the W5500 in another design but I will probably evaluate it first.
Hi JB, welcome to the Prop forum although it seems you are not new to the Propeller obviously. Been lurking hey? Well done with writing the drivers but I couldn't see where it is serving up any web page files?
How did you find the W5500 chip to work with? If you have worked on the W5200 then how does it compare? I'm looking at using the W5500 in another design but I will probably evaluate it first.
Thank you for welcoming me.
Actually I work for Wiznet. That's why I ported the W5500 driver to Propeller.
The demo file is a simple webserver and just send the HTTP reply "Hello World!" characters stored in the code as below.
Thank you for welcoming me.
Actually I work for Wiznet. That's why I ported the W5500 driver to Propeller.
The demo file is a simple webserver and just send the HTTP reply "Hello World!" characters stored in the code as below.
So can you tell me how this 5500 compares with the 5200? There are so many little quirks with the 5200 and I know that the 5500 receive buffers are directly writable not that that is something I worry about but the annoyance I have is the unreadable nature of the socket tx/rx read and write registers and the fact that they do not even initialize to zero or a known point either. The only way I can really tell at present is if I get hold of a 5500 and actually try it.
So can you tell me how this 5500 compares with the 5200? There are so many little quirks with the 5200 and I know that the 5500 receive buffers are directly writable not that that is something I worry about but the annoyance I have is the unreadable nature of the socket tx/rx read and write registers and the fact that they do not even initialize to zero or a known point either. The only way I can really tell at present is if I get hold of a 5500 and actually try it.
I have attached some images of the first Spinneret Web Server 2.0 prototypes. They are fresh off the production line, so I haven't tested them yet, but I'll keep you posted.
I have attached some images of the first Spinneret Web Server 2.0 prototypes. They are fresh off the production line, so I haven't tested them yet, but I'll keep you posted.
David Carrier
Parallax Inc.
I want one now David! Even untested as I will certainly test them.
I have attached some images of the first Spinneret Web Server 2.0 prototypes. They are fresh off the production line, so I haven't tested them yet, but I'll keep you posted.
I'm excited to get some of these. I've got six live Spinneret-using devices out in the field, all doing fantastically well (well, except for an issue with SNTP, but...). With this new configuration and with more i/o pins available, I might be able to run these devices entirely from the spinneret without needing a second processor. Count me in for a few when they come available!
So from the comparison table the W5500 does not support auto MDIX but the W5200 does. That's disappointing. Older PCs/NICs and many embedded Ethernet capable boards do not support MDIX, so this will mean either having a hub/switch on-hand or having a cross-over cable for direct connection. What was the motivation for removing MDIX on the W5500?
Comments
— David Carrier
Parallax Inc.
Have you decided on a size / part number for the QuadSPI flash device yet ?
Does the command work in std spi mode ?
Maybe they limited the erase commands to be via Std SPI only, to protect them a little from the efforts to streamline QSPI ?
Oh, so it is a command issue, more than a QuadSPI one ? It is 30 seconds, in either mode ?
The Winbond data I have 25Q128 mentions 25 sec typ as a chip erase time - does it erase everything ?.
BTW.2 - OT - Are you still using the W5100? After sorting out a few things with the W5200 I'm much happier with this chip.
I see the RX/TX buffer chip is gone, which means it won't spit garbage to the serial port
Target EEPROM size?
Target flash size?
Which RTC device?
Any chance this will be offered without the 40x header so we can make stackables?
Why don't you use thicker power traces (.02"+)? I've never understood why the Quickstart has tiny power traces.
Nice board over-all.
Of course, my personal preference would be two 2mm daughterboard connectors like the one on the original Spinneret. That would leverage Parallax's investment in that format and provide plug-in compatibility with the boards that already use it.
-Phil
True, but I suspect the QS pinout is rather locked-in now.
What about a Raspberry pi connector in a corner ?
If there is not room for a Raspberry Pi 26 pin P1, then maybe a 10 pin P5 ?
That has i2c or i2s options, and i2s on the Pi looks to have quite good HW support. (duplex FIFO depth, at least)
I'm not sure of the drivers, or if i2s can be used as pure digital IO transport (a RPi expert may know ?) , but the stated 3.072MHz looks (just?) do-able on a Prop ?
3MBd duplex (or even half duplex) is quite a decent slave link, all on 10 pins .
Looks like what I predicted and what I was hoping for but didn't say (the flash memory and sd card slot). I had assumed a bunch of the pins would be used already, but I figured some QS addons might work if they didn't overlap. The QS connector could be reversed on this board so that QS addons hang off the edge instead of overlapping... At least that way someone could make an addon board that would work on either this or the QS (assuming they only used P0-15).
Anyway, really happy it has the W5500. Looking forward to getting one when they are available!
Does this mean Parallax will start stocking the W5500? That would make parallax a one-stop-shop for a project I'm working on now.
I made a W5500 driver for Parallax and open it on my github. https://github.com/jbkim/Parallax_W5500
It’s based on Mike’s code but I don’t use the Asm because of better understanding about the SPI operation.
-JB
I'm glad I'm not the only person with this opinion. I much prefer the Propeller Platform form factor over the Quick Start pin layout.
Hi JB, welcome to the Prop forum although it seems you are not new to the Propeller obviously. Been lurking hey? Well done with writing the drivers but I couldn't see where it is serving up any web page files?
How did you find the W5500 chip to work with? If you have worked on the W5200 then how does it compare? I'm looking at using the W5500 in another design but I will probably evaluate it first.
Thank you for welcoming me.
Actually I work for Wiznet.
The demo file is a simple webserver and just send the HTTP reply "Hello World!" characters stored in the code as below.
DAT
index byte "HTTP/1.1 200 OK", CR, LF, {
} "Content-Type: text/html", CR, LF, CR, LF, {
} "Hello World!", CR, LF, $0
So can you tell me how this 5500 compares with the 5200? There are so many little quirks with the 5200 and I know that the 5500 receive buffers are directly writable not that that is something I worry about but the annoyance I have is the unreadable nature of the socket tx/rx read and write registers and the fact that they do not even initialize to zero or a known point either. The only way I can really tell at present is if I get hold of a 5500 and actually try it.
Try it and let me know it if you have any problem.
David Carrier
Parallax Inc.