tag:blogger.com,1999:blog-19092890.comments2026-10-14T19:52:03.559+03:00Vladimir Koifmanhttp://www.blogger.com/profile/01800020176563544699noreply@blogger.comBlogger22273125tag:blogger.com,1999:blog-19092890.post-65609426610938124142026-10-14T07:37:24.574+03:002026-10-14T07:37:24.574+03:00why always so defensive...Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-19092890.post-74906513073295782962026-10-14T04:42:21.923+03:002026-10-14T04:42:21.923+03:00I’m afraid you forgot to end your statement by saying “thanks for your attention to this matter!” Lol Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-19092890.post-87182392704860104152026-10-13T19:01:15.250+03:002026-10-13T19:01:15.250+03:00It’s AI not SI.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-19092890.post-37469271941971852002026-10-13T13:34:30.733+03:002026-10-13T13:34:30.733+03:00A couple of these anonymous comments remind me of the 90&#39;s, except there was no ISW Blog in those days. But, it is a sad day when one cannot publish papers to introduce concepts, whether smart imaging via focal plane image processing, challenging CCD dominance, proposing imaging by counting photons, or in this case, using a CCD in a CMOS process to deliver charge packets to different amplifiers in a pixel, depending on the desired imaging modality. Think about it, edge sensing for AI/SI is a lot more useful if SI can switch imaging modality in a programmable way on the fly. The cost is obviously fill factor today, but maybe not tomorrow. There are other applications too, where such a P.R.O. concept might be helpful and cost-effective.<br /><br />I am proud of my students for their hard work bringing this concept to life, for demonstrating that functionality on first silicon, and sharing our ideas with the community. As an image sensor community, we need to be supportive of next-gen student research exploration and not make discouraging remarks. And besides, quantitative characterization, since the subject paper was submitted, has been promising and will likely be reported in a forthcoming more focused paper - less on concepts, more on experimental results. That&#39;s the way new ideas come to life or lead to better concepts later.<br /><br />As a final comment, I note that IISW came into being 40 years ago to fight against the culture that only finished products should be presented at conferences and in papers, and that work in progress is unworthy of publication. It has worked out pretty well for our community.Eric R Fossumhttps://www.blogger.com/profile/09413387887974939684noreply@blogger.comtag:blogger.com,1999:blog-19092890.post-77609495136075199012026-10-13T02:41:20.824+03:002026-10-13T02:41:20.824+03:00Perhaps wait and publish in a worthy journal when you have the measured results. It just doesn’t look like to be a serious work or paper!Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-19092890.post-31948002771800488722026-10-11T18:35:17.117+03:002026-10-11T18:35:17.117+03:00The building blocks are indeed old (floating-gate resampling dates to 1975, the Skipper CCD to 1990), and so far this is a concept demo: noise, charge transfer efficiency and dark current come only from TCAD/SPICE predictions, and the images are qualitative. What’s new is run-time routing of the same charge packet through an in-pixel buried-channel CCD to either a fast floating-diffusion readout or a non-destructive Skipper floating-gate readout, targeting &lt;0.5 e⁻ rms within 50 samples at room temperature in a 180 nm process (vs. 0.15 e⁻ after 3025 samples at 118 K in prior Skipper-in-CMOS), so the promised characterization will decide whether this matters.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-19092890.post-78585317583554949702026-10-11T03:58:15.457+03:002026-10-11T03:58:15.457+03:00Feels like a paper from 1990s. Lol Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-19092890.post-26190040604381834542026-10-09T16:43:32.079+03:002026-10-09T16:43:32.079+03:00Industrial camera on a chip. Not much left to differentiate for the industrial camera manufacturers of the future.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-19092890.post-22641199260836724052026-10-02T09:26:29.369+03:002026-10-02T09:26:29.369+03:00we hope development QIS， please provide advise.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-19092890.post-78479968762439238762026-09-26T12:59:22.566+03:002026-09-26T12:59:22.566+03:00Congratulations on this prestigious honor Robert! One is selected and elected for Academy service due to their demonstrated expertise and thus ability to provide advice to the government and other agencies, and serve the public good. It is really all about service but it is still quite an honor to be selected and elected. We are all proud of you Robert and also greatly appreciate your service to the IISS.Eric R Fossumhttps://www.blogger.com/profile/09413387887974939684noreply@blogger.comtag:blogger.com,1999:blog-19092890.post-65257136457446305672026-09-23T11:20:21.828+03:002026-09-23T11:20:21.828+03:00Great article for LOFIC introduction!Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-19092890.post-78338280644552943422026-09-09T10:46:41.314+03:002026-09-09T10:46:41.314+03:00Good guestAnonymousnoreply@blogger.comtag:blogger.com,1999:blog-19092890.post-61495883510959498392026-09-06T15:24:18.614+03:002026-09-06T15:24:18.614+03:00I guess they will implement multi-in-pixel memory using ultra-high-density 3D-MIM capacitors in a voltage-domain global shutter pixel, combined with a stacked CNN core.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-19092890.post-58716129907855975302026-08-27T14:57:16.740+03:002026-08-27T14:57:16.740+03:00Sounds like Sony-Honda JV on EV..Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-19092890.post-49532980761509951192026-08-24T21:45:03.017+03:002026-08-24T21:45:03.017+03:00How is this achieved? With gated exposures? Event based SPAD sensors are very bad at dynamic range.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-19092890.post-46442693044334847482026-08-20T05:18:43.794+03:002026-08-20T05:18:43.794+03:00Congratulations, Prof. Henderson.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-19092890.post-550572393891619862026-08-18T10:51:25.033+03:002026-08-18T10:51:25.033+03:00Said well professor! AmenAnonymousnoreply@blogger.comtag:blogger.com,1999:blog-19092890.post-49996408989136006592026-08-17T18:43:05.512+03:002026-08-17T18:43:05.512+03:00In my view, Indie Semiconductor is not particularly interested in the broader product portfolio, with the possible exception of the Mira sensor family.<br /><br />It is possible that Indie Semiconductor will face the same challenge that ams encountered in the past: acquiring a number of second-tier companies does not automatically create a Champions League player.<br /><br />Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-19092890.post-1981700226809239652026-08-17T15:44:11.909+03:002026-08-17T15:44:11.909+03:00who speaks of custom cis development offering? if you just want to buy one of the CMVs because you have designed it in a long running applications... will you be able to buy? they do not even seen to answer requests. strange situation.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-19092890.post-27365308676992448712026-08-17T10:34:40.659+03:002026-08-17T10:34:40.659+03:00This was so long overdue! Very happy to see it and I hope we can finally leave &quot;lux&quot; behind in the camera/LiDAR worldAnonymousnoreply@blogger.comtag:blogger.com,1999:blog-19092890.post-70353752997024923982026-08-14T11:20:00.499+03:002026-08-14T11:20:00.499+03:00​If a multi-billion-euro sensor giant like ams OSRAM couldn&#39;t make small-to-medium volume custom CIS accretive after ten years of investment, indie faces a steep uphill climb, whishful thinking.<br />​Custom image sensors are an inherently high-NRE. The moment a company tries to run a CIS design group in &quot;maintenance mode&quot; to save cash, the technology falls behind market requirements within two silicon cycles.You end up having to sell the business.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-19092890.post-26820215055600845972026-08-10T07:33:55.439+03:002026-08-10T07:33:55.439+03:00Would need to use long enough exposure time to capture traffic lights without flicker (&gt;=11ms).EEngineerhttps://www.blogger.com/profile/15180414668962691783noreply@blogger.comtag:blogger.com,1999:blog-19092890.post-88596763370422645002026-08-08T11:21:36.589+03:002026-08-08T11:21:36.589+03:00The Belgium/Antwerp image sensor ecosystem has a long history of working under the umbrella of an American mother company. <br /><br />The story usually doesn&#39;t end well.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-19092890.post-92015553240776870692026-08-08T11:08:56.175+03:002026-08-08T11:08:56.175+03:00Tetra-Pixel technology with it&#39;s 16 same color pixel clusters is a plague that needs to be eradicated for the good of humanity!Enohttps://www.dpreview.com/noreply@blogger.comtag:blogger.com,1999:blog-19092890.post-11609052717275233772026-08-08T11:07:49.109+03:002026-08-08T11:07:49.109+03:00Tetra-Pixel technology with it&#39;s 16 same color pixel clusters is a plague that needs to be eradicated for the good of humanity!Anonymousnoreply@blogger.com