Practical experiences with standard-cell based datapath design tools. Do we really need regular layouts?
P. Ienne, Alexander Grießing
Abstract
P. Ienne, Alexander Grießing
Abstract
Commercial tools for standard-cell based datapath design are here classed according to design flows, and the advantages of each class are discussed with the results of two test circuits. Algorithmic generation of netlists and of relative cell placement can help reducing area but, contrary to common belief appears often detrimental to speed. Extraction of regularity from synthesized netlists is difficult and requires counterproductive simplifications to the synthesis process. Most promising are synthesis tools which can generate placement data; yet, no tool of this class appears ready today.
OpenAlex reports 2 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
Commercial tools for standard-cell based datapath design are here classed according to design flows, and the advantages of each class are discussed with the results of two test circuits. Algorithmic generation of netlists and of relative cell placement can help reducing area but, contrary to common belief appears often detrimental to speed. Extraction of regularity from synthesized netlists is difficult and requires counterproductive simplifications to the synthesis process. Most promising are synthesis tools which can generate placement data; yet, no tool of this class appears ready today.
Key concepts: Datapath, Computer science, Class (philosophy), Standard cell, Physical design, Process (computing), Computer architecture, Integrated circuit layout