| ▲ | Ultrafast Optical Detector(tdk.com) |
| 32 points by JohannMac 4 days ago | 13 comments |
| |
|
| ▲ | mmmBacon 2 days ago | parent | next [-] |
| While interesting this approach is not any faster than the Ge based photodiodes that we have currently. State of the art is ~-110GHz in Germanium Si photonics using standard fabs. This is only 50GHz. |
| |
| ▲ | glumreaper a day ago | parent | next [-] | | The 50GHz or 20ps figure is actually the limit of the gear the scientists had available - from the paper at https://iopscience.iop.org/article/10.1088/1361-6463/ad9284/... "the signal state transitions were established at less than 20 ps, which is comparable to the bandwidth limit of the oscilloscope utilized at 50 GHz. [...] the initial rise time constant may be shorter than the measurement limits of the oscilloscope." | | |
| ▲ | mmmBacon 20 hours ago | parent [-] | | Sure. But it’s on them to prove their claim that the device is faster and they haven’t done that. There are faster scopes and you can rent them. |
| |
| ▲ | momoschili 2 days ago | parent | prev | next [-] | | A key here for western users is that it does not include Ge, which is notably a material that China has a stranglehold on. | | |
| ▲ | jiggawatts 13 hours ago | parent [-] | | Canada and Spain are major producers. America will be fine as long as they don’t start a trade war with every country all at once. Oh.. ohhh… you threatened Canada with annexation too? Oh boy. |
| |
| ▲ | glumreaper a day ago | parent | prev | next [-] | | I got the impression that "faster" refers to potential bandwidth, as the device responds to shorter wavelengths (300nm to 1.6nm) than other photodetectors. Shorter wavelengths means higher frequencies, "number go up" = more data per second. This is press release language though, it could mean anything. | |
| ▲ | frainfreeze 2 days ago | parent | prev [-] | | Isn't state of the art already in the 300Ghz range? | | |
| ▲ | mmmBacon 20 hours ago | parent [-] | | Sure for InGaAs. The paper is claiming faster for AI so the relevant metric is to compare to what can be shipped today for such applications. To scale for AI we need devices that integrate with standard CMOS processes. More exotic materials can be integrated but they are not very practical for highly integrated systems. NVIDIA’s recent announcement for their copackaged optical switch is in a silicon only ecosystem. |
|
|
|
| ▲ | davesque 2 days ago | parent | prev | next [-] |
| This part struck me: "boosts data transmission and data processing speed, especially in AI applications" How is this better for "AI" than it would be for any other general computing task? Looks like the AI-crazed execs had a go at this press release. |
| |
| ▲ | nashashmi 2 days ago | parent | next [-] | | Sometimes EXECS want AI oriented products. So every product development team adds that their product serve AI too. Just to get attention to their product . | |
| ▲ | ChrisMarshallNY 2 days ago | parent | prev | next [-] | | Probably the case, but I'm not losing sleep over it. Wasn't that long ago, when I would see the word "blockchain," at least once, in every pitch deck. | |
| ▲ | momoschili 2 days ago | parent | prev [-] | | what other general computing task leverages the wide bandwidths that optical transceivers provide as well as AI? I get that AI washing is annoying, but this seems like such a petty thing to be mad at. |
|
|
| ▲ | JohannMac 4 days ago | parent | prev [-] |
| Novel approach to optical detection which is faster than current approaches and can be incorporated on an IC. |
