Jump to content
Search In
  • More options...
Find results that contain...
Find results in...


  • Content Count

  • Joined

  • Last visited

Everything posted by leadeater

  1. The maximum power a 10k RPM 2.5" SAS disk will use is ~9W (8.8W for the one in my hand). Average will be lower than this as you won't be 100% utilizing it all the time but 5W-6W is a good figure to use, a 3.5" 10k RPM SAS disk will use more than this at 15.75W maximum. I don't think I have any 15k RPM disks on hand right now but it's probably safe to just add 20% on to these, if I remember I'll update tomorrow when I'm at work and can look.
  2. And if they did so that may will just start another lawsuit due to them ceasing operation in the EU affecting millions of EU citizens and I think the EU might have something to say about that.
  3. That's not how that technology works neither can it directly increase performance. Increasing data throughput in a system will not increase performance if the computational resources do not scale with it. Also the technology you are talking about requires specific support in the game, GPU drivers, DirectX feature support, and OS support does not enable it in games that were made before it existed. This isn't any different to games being made with DirectX 11.1 feature set not supporting DirectX 11.2 features, everything including the game itself has to support and utilize it.
  4. Not every site can actually easily dynamically scale, things like database caching services can only accept so many connections and automatically adding servers for that may not be possible or necessarily safe to do without oversite. The simpler the website the easier it is to just keep scaling out capacity, once you start adding more complex functions it gets harder and harder. Auto scaling web services is more of a aspirational goal for most corporates, they don't actually have them or there are technical limitations to which parts and components of the site can actually be scale
  5. Well that mostly depends on how the website is architected and where the load demand is highest or if increasing front end web capacity will just overload other components like a database server. If the load is mostly just people viewing products pages etc then that is a fair decent use case for putting that out to a CDN.
  6. I'm well aware of the cost of them, also there are very many very good cost effective options. You don't actually have to spend as much as many do and a lot over purchase for their actual needs. I actually do have first hand experience with studio recording and also live event production, most of it before the digital switch over but some during that time. And the point was if you can spend that much on a Mac Pro you can spend that much and more updating your equipment. As I said the audio community is rife with hold outs and stubbornness to continue doing it they way they always h
  7. When you can get 256 channels (128 in & 128 out) on a single PCIe interface pretty well the only people affected will be those willing to spend $20k-$40k on a new Mac Pro but can't be assed investing in modern equipment and methods, which I will admit is part and parcel with the audio community so Lets also not get too lost here, you can achieve a hell of a lot with a MacBook Pro or a Mac Mini even in relatively high end/large recording studio situations. I'm really not seeing a need for 5 PCIe cards for these people unless they are the aforementioned hold outs
  8. Cooler goes above the height of the PCIe bracket, yea that's not actually going to be useful for many servers at all.
  9. Everything other than GPUs is single slot so it's really not that much of a problem unless you want multiple GPUs. The current Mac Pro case is nice, but it doesn't actually need to be that large. The spaces between a lot of the PCIe slots isn't really necessary, it's nice to have but that's about it honestly. I think the size of it has more to do with a statement and style piece than anything. I mean I do really like it but I can see a lot of wasted space so if Apple does go down the route of making another smaller Mac Pro case I'm pretty sure Apple themselves will start to question if they st
  10. Supporting multiple models is also expensive, more so if you don't actually need more. It depends what the yields actually are and 3 models could actually be more than enough for this. Very few CPU models from either Intel or AMD actually exist because of binning, as in they have so many failing they need these extra models. Other than the very high end which are the most likely to fail the rest are all very much artificial. I mean do you really think Intel actually needs 11 16 core SKUs on a die with 28 cores for binning reasons?
  11. BGA actually makes that issue easier not harder, also there isn't really any difference between soldering the LGA socket to the board versus a BGA chip, just slightly easier with a pick and place with a BGA chip than an LGA socket is all. As for existing products Intel Cascade Lake-AP are BGA, FCBGA-5903 (BGA). https://en.wikichip.org/wiki/intel/cores/cascade_lake_ap As for why Apple might actually do BGA, that's just what they do now with many of their other products and I personally think Apple will offer a maximum of 3 CPU options and I wouldn't be surprised if
  12. Shouldn't be, that server should catch most of it on the front with the filtering it has anyway. Either case it's easy to clean these anyway so if you notice the average fan RPM increasing just open it and clean.
  13. Some ISP's restricted inbound port access to commonly abused ports like 21, 22, 25, 3389 etc so it may be a case of your ISP blocking it before it even gets to your router. Try setting up a port forward rule that maps an external port number above 1023 to the internal port number of 22 i.e. [external IP]:2222 to [internal IP]:22. Then all you need to do is point your application to the external IP and use port 2222 in the connection settings. If it's working internally you can pretty much rule out the server being the issue which means it has to be related to port forwarding config
  14. So exactly what I said, you want to see that there are issues so there are issues Firstly the article is pointless and itself proves how not 'regular' they are, limited in scope, or didn't even cause data loss at all with some just simply moving files. Of that entire thing the single and only one it mentioned that actually has loss of data is the one from 2018, the other actually in existence one came after the article so I don't expect clairvoyance. These articles exist for the same reason you keep posting about them and make these 'regular' not regular claims, d
  15. Except it's not regular and nether wide spread and actually applicable to all users, none so far have been that. One was a bug in chkdsk on specific platform and SSD models and the other was only if you had configured a non standard redirection setting for some of your User folders. So it's neither regular and certainly not wide spread. There's been more data corruption issues in BTRFS and those didn't require specific hardware nor specific configurations outside of normal usage, the only people that ever talked about these were people actually impacted by it, weird huh
  16. Pretty sure majority still do, even if it's just to download them. We actually use SCCM which leverages WSUS to get the catalogue information but we simply do not allow servers to just go out and get Windows Updates directly from Microsoft. Regardless without WSUS (or Windows Update for Business, not actually used it nor want to) you cannot pick and choose which updates you get if you go direct to Microsoft, it's not supported anymore. Tools such as SCCM just sync the Windows Update catalogue to get the KB direct download link and then download them and manually install, however th
  17. Well with the current Mac Pro you can do CPU swaps and RAM swaps, into another Mac Pro or to a completely non Mac device which is nice if at some point you come across a now unwanted Mac Pro and you want the CPUs and RAM for a rackmount server, or the reverse and you want to take it out of a rackmount server and put it in to a Mac Pro. These possibilities become closed moving to Apple SoCs, I hope the bigger ones are socketed but I have a feeling Apple will not go that path and will do BGA instead and you have to buy the right CPU from the start otherwise RIP. *cough* I told you
  18. Coming from someone who didn't bother to read that this entire story is not correct at all holds about as credibility and a colander holds water. Again the issues just aren't as bad as you want to make out.
  19. Yea pretty much, same for most of the past ones too but that's how they survive so I don't really care much about them doing it. Well no not really, this is another falsehood driven by bad reporting. Microsoft has already spoken to this and actually detailed the restructure. They moved many of the QA team in to the development team and changed their workflow methods to 'more modern' methods of continuous development and integration with a lot more tools for automated test plans. Everyone jumps on this "they got rid of the QA team" train without actually looking in to the details
  20. Well correct me if this is wrong but wasn't that once in 25-30 years of chkdsk existing and was specific to certain hardware platforms and SSDs so was not a "this will corrupt your data 100%" issue? What I think is equally shameful is people purposefully seeking out every issue possible no matter how big or small, ignoring the details or if its even valid at all to have a meaningless dig. For all you and others make of it I and others simply do not experience the scale of issues you try and make out, far from. Sure I get not liking something, nobody has to actually like Windows 10
  21. Well except it seems it doesn't so Everyone in the article comments that tested could not reproduce any data loss or corruption, same is true for the people of this forum that tested it. Just because something throws an error and says there is corruption doesn't actually mean there is, we all know how notoriously bad Windows error messages and application error messages in general can be. Like Windows can throw access denied or file not found errors just from file paths being too long within explorer and neither of those errors are actually correct.