Skip to content

Review of Guideline #27

Magnus Jahre edited this page Aug 30, 2018 · 2 revisions

Review of Guideline #27

Formatting

Item Outcome
Guideline complies with the Guideline-Template Yes
Name of guideline responsible with affiliation is clearly stated Yes

Is the audience of the guideline correctly specified?

Yes.

Is the expertise required to write the guideline correctly specified?

Yes.

Are the keywords well selected?

Yes.

Are parts of the guideline too generic or too specific?

The guideline is too specific. Some FPGAs have hard floating point units, and for such devices the advice does not necessarily hold. I say necessarily, because the number of floating point units is limited so you might have problems for these devices as well.

To alleviate this issue, I suggest renaming to "Avoid floating point computation on FPGAs without hard floating point units". The Zynq seems to not have such units -- so the advice holds for the TULIPP platform -- but others do, see https://www.intel.com/content/dam/altera-www/global/en_US/pdfs/literature/po/bg-floating-point-fpga.pdf.

Does the guideline explicitly refer to the handbook? To which part of the deliverable is it relevant (e.g., chapter of D1.2/D1.3)?

It is relevant to the hardware chapter. If floating point is necessary, you should probably select an FPGA with hard floating point units.

Does the guideline specify the work done in the project that can benefit from the guideline?

Yes, but it is bit unclear. The link to the project is made in the insights section, but should be made in the instantiation section.

Other comments

  1. Guideline advice: OK, but imprecise. Broaden to cover FPGAs with hard floating point units.

  2. Insights that led to the guideline: OK, but a bit high level. Should be elaborated a bit such that non-experts can understand.

  3. Recommended implementation method of the guideline along with a solid motivation for the recommendation: OK, but a bit short. What are the options available to the developer? Which of these did you choose and why?

  4. Instantiation of the recommended implementation method in the reference platform: Not done yet. How did you implement the recommendation in the SGM algorithm?

  5. Evaluation of the guideline in reference applications: Not done yet. A quantitative comparison in the context of the SGM algorithm and the UAV use case is probably needed.

  6. References: OK. Possibly reference the link above.

Track changes:

  1. 30/07/2018: Made some formatting changes in the guidelines to cope with template.
  2. 30/08/2018: Guideline reviewed.
Clone this wiki locally