Malcolm, using the W5300 is really easy since it is a TCP stack and not an ethernet interface. Essentially you write high level command bytes to the command register (e.g. "open", "close", "send") and read/write data from/to the w5300 memory buffers. The w3500 takes care of all the rest.
On Tue, Apr 15, 2014 at 5:55 PM, Malcolm Lear <malc...@essex.ac.uk> wrote: > I'd agree the physical package should be fine, I've even hand soldered those > without problems. If the Wiznet really has a socket level interface I'd > seriously consider it. I'd have thought just about all the QL's resources > would be used up doing what the W5300 co-processor does and may be a reason > an ethernet interface hasn't appeared before. Although the code argument is > persuasive is it not true that a large part of that codes functionality > would be performed by the W5300. > > > > On 15/04/2014 15:09, Dave Park wrote: >> >> Yes, Malcolm, the interface is at the socket level. >> >> Both of the devices have some pros and cons. I am still open to >> suggestions >> for other options. >> >> One thing I am not concerned about is the pitch and number of pins on the >> WS5300 or any other device. I don't hand solder SMD boards. I use a solder >> mask to apply paste, position the devices then flow them in an oven. I >> have >> done a fair amount of QFP100 and 1mm BGA work with no problems. >> >> Although the WS5300 is technically superior as a device (it is more >> modern, >> it acts as an ethernet co-processor) the work that has already been done >> with the CS8900A is very important. Without the work being done on >> drivers, >> extensions for languages, etc, it doesn't matter how fly the device is. >> The >> CS8900A has some support already. >> >> Peter, could you outline what is available for the CS8900A, please? >> >> This is why I have opened up the discussion. I dread these discussions :) >> but I am sat here with two equally good choices for two very different >> reasons. The WS5300 is a very persuasive CHIP, and the CS8900A has some >> very persuasive CODE. Since CODE is such a big problem... >> >> If the CS8900A is chosen, we'll have a less capable or future-proof device >> that has much work already done. If we choose the WS5300, it maybe has an >> extra ten years of useful life but we have to start from scratch with >> supporting code. >> >> That's the balance. >> >> >> On Tue, Apr 15, 2014 at 3:48 AM, Malcolm Lear <malc...@essex.ac.uk> wrote: >> >>> Hi, >>> I would think the most important criteria would be to choose a solution >>> that reduces the software load on 68K. It seems after a brief look at the >>> data sheets that the W5300's interface is at socket level, is that right? >>> Malcolm >>> >>> >>> >>> On 14/04/2014 22:39, Dave Park wrote: >>> >>>> I have been presented with two good contenders for an ethernet solution >>>> for >>>> the QL. There might be others; I am open to suggestions and may add more >>>> to >>>> the list. >>>> >>>> I would like there to be a *polite, friendly, respectful, >>>> constructive*debate about the relative merits of the two devices. Once >>>> >>>> you have >>>> expressed a view, that's enough. No need to restate it, unless you add >>>> something further. It is very important that discourse be *focused >>>> entirely*on solving the problem of getting ethernet onto the QL. >>>> >>>> >>>> Please read the data sheets and form an opinion about which would >>>> perform >>>> better on a QL at 7.5MHz or a SGC system at 25 MHz. There is already a >>>> TCP >>>> stack in existence; whether you wish to use it or not is a matter for >>>> you >>>> - >>>> one of the chips is a microcontroller that has its own stack. >>>> >>>> Here are the contenders: >>>> >>>> Wiznet WS5300 >>>> http://www.wiznet.co.kr/UpLoad_Files/Re ... _V128E.pdf >>>> >>>> versus >>>> Cirrus Logic CS8900 >>>> http://www.cirrus.com/en/pubs/proDatash ... 00A_F5.pdf >>>> >>>> If there is a clear consensus, I will assemble some functional prototype >>>> boards using the favored device, then send boards and documentation out >>>> to >>>> a limited number of people who express an interest in doing development >>>> work with them. >>>> >>>> There would be some gentle conditions: >>>> >>>> 1. You'd join a developer mailing list and report occasionally on >>>> progress >>>> you'd made to the other three people and me. Others could join the list >>>> to >>>> provide feedback, etc. The list would allow you to self-co-ordinate your >>>> efforts within the group. I would not be the boss of you. I would not >>>> own >>>> your work. There is no schedule or deadline. >>>> 2. Anything you do, once reaching a state of development greater than >>>> "alpha", would be open source and freely distributable. A GIT repository >>>> or >>>> similar would be nice. >>>> 3. I would, upon release, host information and downloads at >>>> SinclairQL.com >>>> so people can explore the code or develop it further. >>>> >>>> If anyone develops anything to a point that it becomes possible to add >>>> ethernet to a future board, or standalone, I would work out a way to >>>> produce them for the community. If not, you can keep the card as a >>>> curio/plaything/collectible, or send it to someone else who has shown an >>>> interest. >>>> >>>> So, let's begin, shall we? >>>> >>>> >>>> >>> _______________________________________________ >>> QL-Users Mailing List >>> http://www.q-v-d.demon.co.uk/smsqe.htm >>> >> >> > > _______________________________________________ > QL-Users Mailing List > http://www.q-v-d.demon.co.uk/smsqe.htm _______________________________________________ QL-Users Mailing List http://www.q-v-d.demon.co.uk/smsqe.htm
