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

PianoPlayer88Key

Member
  • Content Count

    1,413
  • Joined

  • Last visited

Awards


This user doesn't have any awards

3 Followers

About PianoPlayer88Key

  • Title
    Veteran
  • Birthday 1981-04-29

Contact Methods

  • Discord
    2 accounts, ran into server limit.
  • Steam
    http://steamcommunity.com/id/tfcwings/
  • Twitch.tv
    pianoplayer88key (not streaming yet)
  • Twitter
    pianoplayr88key
  • Website URL

Profile Information

  • Gender
    Male
  • Location
    San Diego county, CA, USA
  • Interests
    Computer hardware, software (games, media production & editing, some others, not so much into office type stuff though), playing piano, listening to a variety of music, photography & videography, other things
  • Occupation
    Currently volunteering at a computer refurbishing warehouse. (Idk if it's similar to FreeGeek.)

System

  • CPU
    Desktop: Intel Core i7-4790K | Laptop: Intel Core i7-6700K
  • Motherboard
    Desktop: ASRock Z97 Extreme6 | Laptop: Clevo P750DM-G
  • RAM
    Desktop: G.Skill Ares Red 32GB (2x 8x2GB) DDR3-1600 CL9 | Laptop: G.Skill Ripjaws 40GB (1x 8GB + 2x16GB) DDR4-2133 CL15 SO-DIMM
  • GPU
    Desktop: EVGA (Nvidia) GeForce GTX 1060 SC 3GB | Laptop: Nvidia GeForce GTX 970M 6GB
  • Case
    Desktop: Rosewill Thor V2 Black | Laptop: Clevo P750DM-G
  • Storage
    Desktop (not all are in system): 80GB WD IDE (WD800BB - dying), 250GB WD IDE (WD2500JB), 2x 750GB WD Black (WD7500AAKS & WD7501AALS), 1TB WD Green (WD10EADS), 2x 1.5TB WD Green (WD15EADS), 2x 2TB WD Green (WD20EADS & WD20EZRX), 3x 4TB HGST NAS (0S03664), 3x 5TB HGST NAS (0S03835), 3x 8TB HGST NAS (0S04012), 256GB Crucial M550 2.5" SSD | Laptop: 250GB Crucial MX200 M.2 60mm, 1.05TB Crucial MX300 2.5", (removed) 2TB Samsung/Seagate Spinpoint M9T 2.5"
  • PSU
    Desktop: Corsair AX760 | Laptop: battery & 230W AC adapter
  • Display(s)
    Desktop: Dell U2414H | Laptop: built-in 1080p IPS G-Sync
  • Cooling
    Desktop: Cooler Master Hyper 212 EVO | Laptop: copper plates + heatpipes + fans
  • Keyboard
    Desktop: Logitech K200 | Laptop: built in
  • Mouse
    Desktop: Logitech G602 | Laptop: Logitech M510
  • Sound
    onboard on both. Laptop mentions Sound Blaster X-Fi on sticker, but there's a RealTek HD Audio Manager icon in my system tray I've noticed sound is cleaner in my laptop than desktop.
  • Operating System
    Both: Windows 10 Pro. Build 1703 as of August 2017.
  • PCPartPicker URL

Recent Profile Visitors

6,136 profile views
  1. Here's some pics of invoices including some video cards / GPUs my family has had over the years. There's several, so I'll put them in a spoiler. I think there are a few missing, and I also didn't include a couple integrated graphics, like the ATI Xpress 1250 on the Gigabyte-GA-MA69G-S3H (Feb 2008, $85) or the HD 4600 on the Intel i7-4790K (Jan 2015, $330). I do remember though, that the GeForce2 MX400 in 2002 was my / our last discrete GPU before the GTX 970M or GTX 1060. Edit: I've also wondered what the 286-10 CPU cost by itself in January 1989. Our next CPU after that was the AMD (486) DX4-120, for which we paid $102 in October 1995. I've also been trying to figure out the performance difference between the two CPUs. I can't find any benchmark sites / articles from back then, like no fathers / grandfathers of the likes of Blender, Cinebench, 7Zip, 3DMark, etc. All I've found is this Wikipedia Instructions Per Second article, which has a table of MIPS - the 286-12 scored 1.28, and the Intel DX4-100 scored 70. Whatever the upgrade in price-to-performance from the 286-10 to the 486DX4-120 was, that's the MINIMUM I'm looking for (I'd like possibly more, I've made posts talking about what I'm looking for with maxed-out-settings HEVC 4K encoding, basically wanting it to be at least as fast as maxed-out mp3 encoding on my current CPU) when I upgrade from my i7-4790K in 2021/2 (likely to AMD on a DDR5 platform that has a lot more PCIe & HSIO bandwidth than the i7-4790K + ASRock Z97 Extreme6 combo). (Several weeks ago or so I had 12 HDDs plugged in doing a DBAN wipe - 10 on the motherboard, 2 on an add-in card. I hit the 1GB/s CPU-to-peripherals bandwidth cap - it originally said it was going to take like almost 6 days I think to wipe the disks (largest was 5TB), and some disks that are capable of doing 150-200 MB/s writes were only doing like 50-60MB/s or so. Once some of the smaller ones (like 750GB to 1.5TB) were done, the others picked up the pace, and the whole thing ended up "only" taking several hours under 3 days.)
  2. Ahh very interesting, @SolarNova. I think we both forgot something. You forgot the GTX 280, and I forgot about the 9800 GTX+. (I think part of my issue is IDK where the cards stack up in each generation's lineup before they went to GT x10-x40 and GTX x50-x90.) I'd also be interested in the pricing of cards even older than the GeForce 2 Ti, even going back to these, if anyone knows.
  3. Totally agree here. I think the launch prices of the GTX 200 series refresh are where prices SHOULD be. GTX 275 = $249 GTX 285 = $359 GTX 295 = $499 Source: TechPowerUp GPU database I think of the "x0 Ti" naming (not the Supers) as being the "successor" to the "x5" naming. Interesting thing - the GTX 280 launched I think about 6 months before the GTX 285, for $649. I'll just say, we really need the AMD RX 6000 series to be another ATI HD 4800 series. I'd also like to see the day when even the top-tier workstation card would be under $1000 - I mean INCLUDING tax & 1-2-day shipping without Prime, not the base price being $999. Or, at least have a big enough boost in gaming performance so they'd be the same FPS/$ as the $200-400 GPUs for those who want to game with them. (Imagine the gaming performance if an Ampere successor to the Quadro RTX 8000 scaled like that...) Take a look at the resolution and VRAM usage in this screenshot of a game which I think came out about 5 years ago on PC, and 7 years ago on console: Yeah true, it's not exactly "LowSpecGamer" settings. For the rest of the settings, open the spoiler.
  4. Hmm... 3080 is supposed to be good with 4K gaming, and has 10GB VRAM. 3090 can apparently do 8K, and has 24GB. Umm, yeah, I'm thinking, that sounds a bit low. My desktop and laptop both have 1920x1080 60Hz displays. Desktop has an i7-4790K and a GTX 1060 3GB, laptop has an i7-6700K and a 6GB GTX 970M. (The "reversal" of pairings is because I built the desktop in January 2015 and used iGPU for a while, then got the laptop in December 2015 with an i3-6100, then got the i7-6700K for the laptop and the GTX 1060 3GB for the desktop in November 2016.) In GTA V, the game wants about 9 GB of VRAM at max settings at 1080p. Proof, with screenshots of the settings: That INCLUDES cranking everything up in the Advanced Graphics menu. (The "Frame Scaling Mode" had the biggest difference - turning that down to its lowest setting would get VRAM usage down to I think 2 GB or somewhere way down there even when I didn't touch anything else.) I don't have video of it, but I actually ran some tests a while back with those settings, by running the in-game benchmark sequence. With the laptop's GTX 970M 6GB, I was getting about 6 fps at the beginning of the benchmark. With the desktop's GTX 1060 3GB, I was getting 0.3 fps, and aborted the benchmark about a minute or two into it. (Which wasn't very far - the actual pace of the game was really slow - was taking a full minute JUST to do that initial fly in *TO* the houses in the first scene.) Interesting thing. I took the 1060 out of the desktop and ran the benchmark on just the integrated graphics. (It was taking 2 GB from system RAM - that's what I took those settings screenshots with.) Remember the 1060 got 0.3 fps? Well, the Intel HD 4600 got 1.8 fps. That's about 6x faster, even though in FireStrike with default settings, the 1060 scores about 10.5K whereas the HD 4600 scores around 700-750 or so, or about 14-15X slower than the 1060 for "normal" things. I wonder why the "slower" HD 4600 with less VRAM would have been faster in a severely VRAM-limited scenario, than the "faster" 1060 with more VRAM? My hunch is it has something to do with the fact that when you're out of VRAM, you have to swap to system RAM.... when using the dGPU you have to go out of the GPU, through the PCIe slot, across the motherboard, to the CPU (I think), to the system RAM, and back again, and since you're using 3x the VRAM the card has, you have to do that swap 3x per frame. OTOH, when using the iGPU, its VRAM *IS* the system RAM, so it HAS no latency when it needs more (as long as you have enough - the desktop has 32GB, the laptop at the time had 40GB and now has 64GB), and can do that swap much, much quicker. Speaking of the laptop, I think its 6GB VRAM (and only needing to swap once per frame instead of twice or 3x like the 3GB 1060) contributed to it getting a significantly higher FPS, even though it's about 1.6X weaker in FIre Strike than the 1060 3GB (laptop has gotten around 6.5K score in that benchmark.) Anyway... GTA V came out on PC about 5 years 4 months ago. I imagine that newer games are even more demanding. People that say a particular new game runs well at 60+ fps at 4K or whatever on a modern high-end GPU ... I kind-of think there's one of two things happening. EIther you're not cranking up absolutely EVERY single graphics option in the game to its absolute max (I don't call it max settings unless that's done), Or, the game doesn't HAVE the option to go to such high settings. (As I hinted at earlier, changing frame scaling, render resolution, etc does count - if that option exists, max settings means that would be maxed out too. If I understand right, upping the resolution with the same settings increases the VRAM requirement by an amount equal to the difference in horizontal times vertical resolution. For example, if 1920 x 1080 (1080p, 2,073,600 pixels) needs 9123 MB (8.9 GiB) of VRAM with those settings, then 3840 x 2160 (4K, 8,294,400 pixels, 4x 1080p) would need 36,492 MB (35.64 GiB) of VRAM, and 7680 x 4320 (8K, 33,177,600 pixels, 16x 1080p) would require 145,968 MB (142.5 GiB) of VRAM. I don't have access to one of them, but sometime I'd like to see someone benchmark GTA V (or a newer more-demanding game) at max settings (like I described above) on something like a Quadro RTX 8000 with 48 GB VRAM, and compare it to the RTX 3090 with 24GB and RTX 3080 with 10GB when they come out, at 4K, max settings. I want to guess that the Quadro would do 4K just fine, like 60+ fps or more (maybe 120 or 144+?), the RTX 3090 at 4K would be similar to my GTX 970M at 1080p, and the RTX 3080 at 4K might be like my GTX 1060 3GB at 1080p. Or is GTA V an outlier and most games don't require 9 GB of VRAM at 1080p when you crank all the settings to the max? Or am I the only one who defines max settings as cranking absolutely EVERYTHING up? BTW I was googling Fortnite and Overwatch when mentioning the render scaling earlier ... Overwatch I think supports up to 200% in the menu, Fortnite up to 100% in the menu, but, apparently Fortnite, IIRC, has text files so you can edit them to be pretty much almost any setting you want. I don't have either of those games (well technically I have Overwatch since Oct 2016 but have never had time to install / play it, also never have downloaded Fortnite) so can't check it out for myself right now, but for the above "max settings" discussion, I'll be a little bit merciful on the PC hardware and say it would only need to be max settings in the game, not going so far as to put ridiculously high custom numbers in text configuration file. Speaking of super high resolution / settings though ... What currently-existing (or soon-to-be-released) GPU would be required to play modern AAA titles ... on one of those several-foot-tall video walls consisting of, for example, 32" 8K monitors edge-to-edge across the ENTIRE wall, at max settings, never having a single frame take longer than 16.66 ms to render?
  5. Yeah. I'm leaning toward that option, I feel like it's easier to just take them all to one place. I have a shared folder on Google Drive with several subfolders - including some SMART info on most drives and benchmarks on a few. (It doesn't include some recently-purchased drives though.) I also have some screenshots, but they're kinda scattered all over the place (some on Google Photos, some in albums, some may still be on local drives) so if I tried linking them people's brains would fry.
  6. I'm working on creating some kind of strategy to get something started with that. If you opened the spoiler, I list all the drives I have, including the drives I'm getting rid of, with format coding for different categories of what I'm doing with them. That includes mentions of some of the larger HDDs I have, with some being used for storage and some for backups. Basically though... for data storage, I have 4x 8TB, 2x 10TB and 2x 14TB drives. Most of my data is on an 8TB and 10TB HDD, with the rest on several SSDs. My current plan is to use 2x 8TB, and 1 each 10TB and 14TB for data storage, and the others for backups of the same. (The data on the SSDs in my laptop would get transferred periodically to one of the "main data storage" drives, then transferred to an equal-size backup drive.) I wonder if I should keep 1 or 2 of the 4TB or 5TB drives, or maybe order another HDD for backing up my SSDs. (If I do, I'd prefer a large enough one so that ALL my SSDs would fit on 1 HDD. I'll skip the math details here but basically that would work out to about 8.85 TB of total SSD capacity for now, with the next size HDD up being 10TB.)
  7. Hey all... So I'm getting rid of about 20 or 25 or so old HDDs of various types and sizes. I'm planning to take them to a local charity / recycler next week, but there's a few drives I'm still trying to decide on and I would like some input on that. The main drives in question are 6 HGST Deskstar NAS drives - 3x 4TB (HDN724040ALE640) and 3x 5TB (HDN726050ALE610). Should I... Include them in the box with the older / smaller drives? Try to sell them? (I don't have a way to ship them, though, and I would like to keep the retail boxes for some of my other drives that are lacking boxes; also if I sold them I'd prefer to sell all at once), or Keep them? (I don't think I need to though, and I'm trying to pare down my number of hard drives. Also a single 14TB I have holds more than 3x 4TB and a bit less than 3x 5TB.) My data for now fits on an 8TB and 10TB HDD, plus about another 2 or 4 TB or so on several SSDs. Newer 3.5" drives I have are 4x 8TB (3 HGST, 1 Toshiba), 2x 10TB HGST, and 2x 14TB Toshiba. (The plan is to split them evenly - half for data, half for backups, hence why I have an even number of each capacity.) Total list of drives in spoiler below, hope I haven't forgotten any. (lol .. I was tempted to convert the capacities to base 2 / 1024, so for example 250GB would be like 233GB, 1TB would be 931GB, 5TB would be 4.55TB, 14TB would be like 12.7TB, etc.) As for retail boxes (with shock-resistant packaging material), I have 2 boxes from older WD drives, as well as boxes from the 4TB through 10TB HGST NAS drives and the 8TB Toshiba N300 drive. Also does anyone know where I could inexpensively get some other kind of container for bulk storage of drives? I'm thinking something similar to these. (In the side by side screenshot, the ProStorage on the right is WAY too expensive (like 4-6X or so), but I'd be willing to pay a little more than the one on the aquaphoenix website on the left ... unless I could get one or a few for free or close to it locally. Bonus points if it's a container that doesn't thermally insulate the drives, so if I wanted to hook up SATA cables and run the drives while they're in it (also be able to secure the vibrating drives to it), it would be fine. (Or would I need something like this to be able to do that?) Also I think I want to get rid of a few other things, wondering if I should take them to the charity / recycler as well, or do something else (possibly including keeping some)? They are Rosewill Thor V2 case (bought 2015) OCZ StealthXStream 500W PSU (2008) 4x 1GB G.Skill DDR2-800 DIMMs (2GB 2008, 2GB 2009) Intel Core i3-6100 CPU w/box & cooler (2015; if I don't build an LGA1151 NAS, for example with an MSI C236A Workstation or something. Couldn't use the Thor V2 cause the HDD bays don't support the wide-screw-hole HDDs.) 1x 8GB G.Skill Ripjaws DDR4-2133 SO-DIMM (2015; was in laptop, so was i3-6100; laptop now has i7-6700K and 4x 16GB SO-DIMM.)
  8. I'm guessing they'd be basing that 90% number on the total number of individual units sold. Whereas... I might think of it more as 90% of the models available for sale recently, with each GPU SKU (like MX330, MX350, GTX 1650 .... RTX 2080 Max-Q, full-fat RTX 2080 Super, plus AMD which I haven't researched as much yet) counting as one. So at whatever point would be the "90%" of GPU models ... I'm guessing somewhere around the RTX 2060 Super or RTX 2070 Max-Q ... I might "expect" Intel TGL Xe to be better than *that*. I'm not in the market for a Tiger Lake laptop, but if I was going for one, I would probably be in for a world of disappointment. Or maybe they could fudge that "90% of discrete graphics" number another way ... by counting ALL discrete graphics in laptops, all the way back to ... when was it laptops first existed, sometime in the 1980s? Or were there any in the 1970s maybe? And, about that "up to 4.8 GHz" clock .... Lemme guess.... That's only on ONE golden-sample CPU... by "golden sample" I mean.... take a SKU from Silicon Lottery that's near the bottom of their stack, but still above "Intel's stock specs". That SL CPU requires a significant overvolt and LN2, when overclocked, to hit a given clock speed (which may be significantly above 4.8 GHz.) This one golden-sample CPU, when overclocked, can hit that "LN2-on-SL-SKU" frequency - at STOCK voltage (1.2V), using ONLY the Intel heatsink supplied with non-K Skylake through Coffee Lake CPUs. Back to stock settings and the "up to 4.8 GHz" ... Maybe it's not even fully 4.8 GHz - more like 4.795 GHz with the fudge factor of having BCLK drift down slightly from 100 MHz, like 99.91 MHz. Also, maybe it ONLY EVER achieves that clock for *ONE* clock cycle in its *ENTIRE* life (so for 1/4795th of a second), on ONE core, when the others aren't just idle, but actually completely turned OFF. Oh, and that fudged "4.8" GHz is done in an environment like this. Joking partially(?) aside .... I'd someday like to see how far people like Linus, Jay, Steve (both of them), Paul, others could push a more "normal" desktop CPU and GPU in that environment - using *ONLY* passive cooling from the environment, even going so far as to take the heatsinks OFF, also no external fans either, except any natural wind blowing.
  9. My probably fairly limited experience with maxed-out RAM is Generally the system will slow to a crawl, Especially when you have fairly little RAM and have pagefile on a hard drive. If your RAM is tightly enough constrained, even pagefile on an SSD won't help. This system would take several MINUTES to respond to some clicks that a "normal" configuration might respond to before the next screen refresh. When this setup hit its physical RAM limit, the screen / app (Chrome, IIRC) went black, and the PC just about froze for a few minutes. It eventually recovered, though. This one wasn't so lucky. This was the last screenshot I was able to get - not 20-30 seconds later it didn't BSOD, it just powered off, then restarted a few seconds later, complete with POST. I think I may have seen it hit 0 RAM free (or close to it) right before it went down, but couldn't get a pic. I even had several hundred GB free on the 1TB 970 Evo that I had assigned the pagefile to, idk why it didn't increase the swapfile further when it needed it.
  10. Ah. Motherboard is an ASRock Z97 Extreme6 with 2 M.2 slots, laptop with other 2 slots is the Clevo P750DM-G. IIRC the 2 screws I have came with the mobo, I pilfered them a while back to install the 2 SSDs in the laptop. Also I was looking online at a few hardware stores, and getting one of those screws is looking very iffy at best. I'm running out of time, I'm getting to the point where..... which one should I run without a screw? (I'm thinking one in the laptop as the cover is pretty close, but how easy is it for a non-screwed-in M.2 SSD to work itself loose?
  11. Ah, 3 or 3.5mm? Or does that not matter? Also what about thread pitch? Also I thought all M.2 SSD screws were standardized "get the Hardware Spec for the screw if you have the manual for your case"...
  12. How would they know - should I take the screw and show it to them? Also where? Here in San Diego we have Home Depot, Lowe's, Dixieline, maybe Ace Hardware but not sure, also might Best Buy, or as empty as they may be, Fry's have something? Micro Center is probably an hour and 40 minutes each way at about 80 mph, and in the time the round trip takes, I'd like to have the data transferred, AND the OS's reinstalled.
  13. as I will be doing some data transfer and OS reinstallation today. What can I do to securely fasten the 3rd SSD? The 2TB P34A80 and 256GB MX200 will be ending up in my laptop, While the 1TB 970 Evo will end up in the desktop. The Silicon Power is the new one, 970 Evo is currently in the laptop. To do data transfer from the Samsung to the S.P., both will be going in the desktop, booted off a 2.5" 256GB Crucial M550. Desktop board is an ASRock Z97 Extreme6, CPU is i7-4790K. I sometimes run without the GPU (and actually have it out now.) Once SU002TBP34A70M28AB has the data from MZ-V7E1T0, the latter will be reformatted and have a fresh install of Windows 10 in the desktop. CT256M550SSD1, which is now dual-boot, will be reformatted and be a dedicated Linux SSD. So... Either where can I get a proper M.2 screw (or 2+) TODAY, this afternoon, or what's another way I can attach the 3rd SSD? Or, if I have no other choice, which one should run screwless?
  14. I thought there were other sources too besides some I've already seen mentioned? Like what were the Asian based ones... Chiphell or Digitimes or an online forum I can't read without a translator ... (I'm getting the name confused in my mind with Alibaba but I know that's not it - I think it may have a "g" in the web address and be ~8-11 characters long but I'm not sure.) As for youtube, AdoredTV and Moore's Law Is Dead are a couple others I watch. Oh, and another couple sites I go on are 209.58.135.68 and 104.27.189.66, the former of which I visit often enough so it shows on my phone's Chrome home page as one of the top 8 sites I visit. (LTT forum has pretty much had a lock on #1 for at least several months, maybe a couple years.) I see some things there, sometimes months before they're on the more mainstream sites or even the manufacturers sites. Oh and what about the sources (like some twitter users, etc) for the above sites?
×