TechXclusives from Xilinx

Hi all,

Does anybody knows where are the web page of "TechXclusives" subject. It seems they've disappeared from Xilinx website.

Thanks.

Reply to
Alain
Loading thread data ...

And there used to be a nice summary page for different categories of app-notes that is now gone. More improvements to the Xilinx web site for us.

John P.

Reply to
johnp

As you noticed, the website is under de-struction and (hopefully) con- struction. You may also have noticed that TechXclusives had been almost dormant for the past year. We intend to remedy this situation, and come up with a different but similar method of publishing short technical notes, publicly readable. Austin, Ken Chapman, and I are very interested in breathing new life into this project. Give us a few weeks... Peter Alfke

Reply to
Peter Alfke

Meanwhile, you can use the wayback machine...

formatting link

HTH., Syms.

Reply to
Symon

Thanks all for your answers (and interresing to know that this project is still alive) and the useful links !

Alain.

Reply to
Alain

Reply to
Peter Alfke

Peter Alfke wrote:

Please suggest to your web developers that they should put technical documentation (such as data sheets, user guides, app notes, ZIP files associated with app notes, TechXclusives, etc.) at permanent URLs that don't get shuffled around and lost whenever the marketing people decide that the site should be redesigned.

I don't think there's any reason that there can't be a portion of the namespace set aside for permanent URLs for such things, and the fancy point-and-drool web pages can ultimately yield links to the documents at those permanent URLs.

Someone well-known in web development (maybe Philip Greenspun?) pointed out years ago that URLs for real documents (not fluff) should not shift around randomly, but should be persistent. There was a paper on it, but I can't find it now. Unfortunately few web developers have taken this to heart.

Thanks, Eric

Reply to
Eric Smith

OK, but that's no excuse for the web developers to throw the files away. Even if new ones weren't being generated, the old ones should remain available. They contained useful material.

Too many people seem to subscribe to the idea that if information is old, it has no value. :-(

Eric

Reply to
Eric Smith

The point has probably been made in various places, but one is:

formatting link

Mike

Reply to
MikeShepherd564

I understand, and I agree somewhat. But we also have to realize that this is a fast-changing technology, and something that was cool and interesting in 2002 is now often obsolete and perhaps even misleading. That's why we want to overhaul the old TechXclusives, throw some away, and bring the others up to snuff. We want to cater to working engineers, not to archaeologists... An overview of FPGAs that stops with Virtex-II is not meaningful anymore, and neither are complex asynchronous FIFO explanations, when you get ready-made designs in the more modern Virtex chips. Sometimes we will annotate the old stories with warnings that there are better ways of designing with the more modern parts. The same considerations should be applied to application notes. Most of them that are 5 years old are more amusing than enlightening. Scary... Peter Alfke

Reply to
Peter Alfke

I've found very useful info that is still relevant but much more than five years old. For example, your asynchronous clock switching circuit, and Rob Weinstein's "flancter". This is exactly the sort of stuff that tends to get lost when someone gets a wild hair to "clean up the web site". Avnet dropped the flancter app note from their site even though Rob tried to push them to keep it. I've just used several flancters in a new design this week, but I had to beg on the newsgroup a few weeks ago to get people to send me a copy of the actual ap note. (Thanks to everyone that helped out!)

So what if 98% of old XL4000 ap notes aren't relevant? It costs almost nothing to keep them archived in a directory somewhere, and the other 2% may contain useful design techniques that haven't been rewritten for the Virtex 5. These old app notes may help engineers create new designs for paying customers.

Since some XL4000 series parts are still in production, there may even be engineers that occasionally have to maintain the designs or migrate them to newer families. Suppose the HDL source for something has an explanation like "see the example on page 12 of XAPP271828", but the app note and associated files are gone. I've been through that several time with other semiconductor vendors.

I've always lamented how difficult it is to find data on old parts. When data sheets started being supplied in PDF form in the late 1980s, one of my friends said that the problem was solved and that the data would be around forever. I immediately disputed that point, and predicted that electronic data would be even more emphemeral than paper. Twenty years of experience confirm that.

Sigh.

Reply to
Eric Smith

Reply to
Peter Alfke

I fully second this. Especially Asian companies are keen on throwing out old datasheets. Back in the old days design companies used to have several cabinets filled with datasheet books. Nowadays they have hard drives full of datasheets. Websites like

formatting link
are very valuable; perhaps someone should start
formatting link

--
Reply to nico@nctdevpuntnl (punt=.)
Bedrijven en winkels vindt U op www.adresboekje.nl
Reply to
Nico Coesel

Peter -

Have you thought about opening the Tech Exclusives up to outside writers? For example, Antti had some nice examples of using SRL16's in interesting ways.

Just a thought...

John Providenza

Reply to
johnp

Reply to
Peter Alfke

Even though some interesting old techniques may seem useless to users of new products with such circuits already built in, the articles remain useful to allow users that need something slightly different (and not available built-in)..

I agree that these articles often need editing or augmenting to inform would-be users of now-available alternatives, and some articles are just too far gone to resurrect.

But many things old become new again eventually... Just look at CORDIC algorithms, for example. And now they are even falling out of favor again in some cases, supplanted by the availability of more multipliers, etc.

I remember when I first started working with FPGA's in the early 90's (schematic capture), and reused techniques I had learned in college for implementing logic functions with multiplexers and decoders, but had not used for years in favor of PALS with SOP implementations.

Andy

Reply to
Andy

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.