Preliminary tLUT value for Cyclone III

[Altera_Forum]

Dear expert, 

 Does anyone know how to find the preliminary tLUT value for Cyclone III from Quartus II Timequest. Can somebody point me where to find the preliminary tLUT value in timequest report?   Thank you very much......

[Altera_Forum]

What's tLUT, the delay through a LUT? If you do report_timing on a path going through a LUT, look at the cell delay? I wouldn't get caught up in individual delay values like that. The delay is highly dependent on which path is used through the LUT, and is still not a constant from LUT to LUT. (They never were, but older devices used to have timing models that were much simpler than the hardware)

[Altera_Forum]

Hi Rysc, 

 Thanks for your quick reply....   If you check the cyclone II datasheet (http://www.altera.com/literature/hb/cyc2/cyc2_cii5v1_01.pdf?gsa_pos=1&wt.oss_r=1&wt.oss=tlut cyclone III), chapter 5-19, the min & max spec tLUT is given. But, this tLUT specification is not listed in Cyclone III datasheet.   So, how do I get this information? Can I estimate from preliminary tLUT value in timequest? If yes, how to do it?   Thanks....

[Altera_Forum]

You'll have to look at the cell delays through a LUT. Be sure to use the Fast Timing Model to get the min. My guess is that it's not in the datasheet anymore(I haven't checked, but assume you have), because this value is generally pretty useless. Just changing inputs on the LUT, which users don't control, can make a big difference. Routing also tends to dominate critical paths, which is also out of user control. Could you explain why you want this and how you're using it, so maybe we can give more suggestions.  

But if you really want to do it, do a 4-input function like an AND gate. Do a setup check in the slow corner to get the longest delay, and a hold check at the fast corner to get the shortest delay. This should be pretty good for finding the boundaries.