INTL 3:770/1 3:770/3   
   REPLYADDR theom+news@chiark.greenend.org.uk   
   REPLYTO 3:770/3.0 UUCP   
   MSGID:  af1f8460   
   REPLY:  d10ddf8a   
   PID: SoupGate-Win32 v1.05   
   Pancho  wrote:   
   > On 10/30/24 08:50, Andy Burns wrote:   
   > > Pancho wrote:   
   > >   
   > >> The official NVMe Pi Hat has been out for months,   
   > >   
   > > Oh, I don't have a Pi5, and though I kept hearing about 3rd party NVMe   
   > > HATs and lack of official one   
   > >   
   >   
   > OK, I see there is a story about rPi launching actual NVMe M.2 SSDs. As   
   > opposed to a hat. I've no idea why they would do that. The obvious   
   > suspicion is cashing in on a brand name.   
   >   
   >    
      
   It may be they are doing it because supply of small-capacity 2230 NVMe is a   
   bit of a minefield.  eg I checked scan.co.uk and the smallest 2230 they have   
   is 512GB.  There are some 256GB sold by Amazon.co.uk which are more   
   expensive (and a few more dispatched by other sellers, of variable   
   trustworthiness).  At least with the RPi brand you know they're compatible,   
   and they seem to be decent value.   
      
   > It's hard to know what is going on with the Raspberry Pi guys, the   
   > RK3588 devices are clearly faster, lower energy, albeit with shit   
   > software support. Who knows what will happen with the next generation   
   > Arm SoCs. I guess maybe Raspberry Pi have a clue, and hence decided to   
   > monetise the brand now, before a new product wipes the floor with them.   
      
   It'll depend on what fab slots they can get.  Not everyone can fab on the   
   latest process, especially with a budget.  Also how much cache they can   
   afford to put on the die.   
      
   RK3588: (64+64+512)*4+(32+32+128)*4+3072 = 6400KiB   
   RPi5  : 512*4+2048 = 4096KiB   
      
   RK3588 also has 4 extra A55 cores which RPi doesn't have, but is more   
   expensive ($100+ for the Banana Pi).   
      
   > >> I guess I should get one, or maybe an alternative. I just bought a   
   > >> NVMe USB enclosure which has appalling performance   
   > >   
   > > Anyway, is it likely the write speeds are faster than the  read speeds?   
   > > I know some enterprise SSDs come in "read mostly" or "write mostly"   
   > > flavours, but for a Pi?   
      
   > Dunno, IOPS doesn't mean a lot to me. As TNP says, maybe a write   
   > operation is to cache, and a read is from main memory.   
   >   
   > On many solid state persistence devices you see a very fast initial   
   > write (presumably to cache) before quickly settling down to a much lower   
   > rate for big files.   
      
   Any decent benchmarking tool should get past the cache to exercise the real   
   storage.   
      
   I think they've got them around the wrong way.  Their ODM Biwin's 2230 has   
   more read than write IOPS:   
   https://droix.co.uk/product/biwin-2230/   
      
   Theo   
      
   --- SoupGate-Win32 v1.05   
    * Origin: Agency HUB, Dunedin - New Zealand | Fido<>Usenet Gateway (3:770/3)   
   SEEN-BY: 10/0 1 90/1 103/705 105/81 106/201 124/5016 128/187 129/305   
   SEEN-BY: 153/757 7715 218/0 1 601 700 840 870 930 220/70 221/1 6 360   
   SEEN-BY: 226/17 30 100 227/114 229/110 111 114 200 206 300 317 400   
   SEEN-BY: 229/426 428 470 550 616 664 700 240/1120 266/512 267/800   
   SEEN-BY: 282/1038 291/111 292/854 301/1 113 812 310/31 320/219 322/757   
   SEEN-BY: 335/364 341/66 342/200 396/45 460/58 633/280 712/848 770/1   
   SEEN-BY: 770/3 100 330 340 772/210 220 230 5020/400 1042 5075/35   
   PATH: 770/3 1 218/840 221/6 301/1 218/700 229/426