Jump to content

mitch_de

Retired
  • Content Count

    3,042
  • Joined

  • Last visited

  • Days Won

    2

mitch_de last won the day on November 9 2015

mitch_de had the most liked content!

About mitch_de

  • Rank
    InsanelyMacaholic

Contact Methods

  • Website URL
    http://

Profile Information

  • Gender
    Male
  • Location
    Stuttgart / Germany

Recent Profile Visitors

17,111 profile views
  1. If you have an AMD CPU, maybe this also has an negative effect on AMD gpu drivers - Chris may know more.
  2. RX 460 : removed AGPM & non Polaris injects. You see in Luxmark Info, that my RX 460 has 16 Compute Units (like RX 560) not 14. By flashing edited ("unlock Shaders) ROM back on the card. I reduced the max GPU clock a bit down from 1212 MHz to 1208 MHz. The 16 vs 14 CUs alone result in about 8-10 % compute boost - so your around 20000 Metal score is good for 14 CUs. I only get around 17000 Metal score (16 CU) without radeonboost. DL: RadeonBoost.kext.zip
  3. Yep, even my RX 460 gets around 23K GeekbenchMetal (with Boost, without 17K) . Your RX 590 should get at least 40K without the RadeomBoost.
  4. Interesting results. But it shows, that the maximum bandwidths are near the same for transfer sizes bytes more than around 20 KB. In your results the smaller transfersize speeds shows much , up to double , difference between with / without booster. I cross tested that on my RX 460 (booster gain 30%!) with an closer look to the smaller transfer sizes, because like you my maximum BAndwidths are also same (less than 2% diff with/ without booster) , nears same as yours. Between runs the smaller transfersizes vary much more than the bigger transfersite bandwidths. For my RX 460 frist run was onyl about 10 MB/S Device to Device Bandwidth, 0 Device(s) Transfer Size (Bytes) Bandwidth(MB/s) 1024 9.2 2048 146.2 Second run and all other runs around 49 MB/S Device to Device Bandwidth, 0 Device(s) Transfer Size (Bytes) Bandwidth(MB/s) 1024 50.4 2048 160.4 So, if you also repeated the test are you sure also Bandwidths of the smaller transfersizes are between the runs same much lower with the booster? If yes, that could explain why booster shows higher Geekbench Scores, they may use not much memory blocks by their around 8 small bench tasks.
  5. Yep, without checking (against other apps than geekbench) it is unclear if the boost 1. is really happen also in other apps Some investigation - i will look after the files (libs, amd drivers etc.) are loaded and used when geekbench is running compute - with boost kext and without. Perhaps thats the difference. Or, perhaps geekbench checks something / optimise something in their app which result in boost. EDIT: Geekbench loads and uses same libs /AMD driver etc. @Xanny: Can you try to measure the bandwidth speed with boost (kext) and without. Perhaps , if you measure higher value for the bandwidth iit may be the reason. For me its not - RX 460 has same bandwidth, doesnt matter using kext or not. Small oclBandwidthTest is in my last post (2 post above this). Your bandwidth will be much hogher than my around 80000. I think at least 120000. But the interest is only in the difference (perhaps) with kext (or device properities inject) and without.
  6. With the kext and my RX 460 , 10.15.4, i also get around 30% higher geekbench 5 OpenCL and Metal scores, So far so good. 1. I don't get any boost in other gpu compute benches like LUXMARK etc. and no boost of FPS in any Game with buildin FPS counter i testet. 2. I can get (Thanks to Chris!) same higher, 30%, Geekbench 5 scores - and only there - if i inject 3 device properties by Clover (Screenshoot) So i get same boost also without the kext. AGPM (is within newer kext versions) has no effect on my RX 460 card. Chris tells that the reason for higher Geekbench scores (some get 50%++!) came by higher vram bandwidth/internal VRAM tranfers and access goes faster and Geekbench is sensitiv to that. I dont think so because Geekbench measures score by about 8 different small gpu compute tasks. Even if some are sensitiv to the vram bandwidth it tells not why other gpu compute tasks or benches show really zero effect - for me Luxmark speed excat same not even 1% difference. And Luxmark uses also big ammount of the VRAM (up to 2 GB) of course. So why most of us - if they have an boost more than 10% between 2-3 geekbench runs - cant measure any boost with other benches or apps which can be users as benches. "FCP runs smoother" is for me or others not reproducible - we would need same source video and must do same things so compare. I tested some games which shows same FPS with/without kext and Luxmark + other gpu compute benches - the 30% Geekbench boost wasnt shown in any, not even 1%. I will NOT say the kext is placebo because Geekbench shows reproducible effect - but perhaps the boost comes by geekbench itself, geekbench optimises some on the fly in their gpu compute code if it found those kext injected parameters? Really, if i get 30% boost with kext, others up to 100%!, but cant find any boost in other benches / apps something is to investigate. Great would be others with same boost whoch find an bench / app whichs shows the booast by values and not "feels smoother, faster". I used also Idigo Render Bench http://downloads.indigorenderer.com/dist/beta/IndigoBenchmark_v4.0.64.pkg, but also here no boost, same as without kext. HINT: Disable CPU as render device Should bei gpu only like on sceenshoot. PS: Maybe oclBandwidthTest is the rigth tool to verify if the these "bandwidth/VRAM speed" may explain why geekbench boosts with the kext. For me the bandwidth is even a little bit lower using the kext (i belive by AGPM content) by using device properties inject bandwidth is same again. With or without device properties (=boost kext) i get same 80 GB/Sec in the oclBandwidthTest. Usage: Start Terminal, oclBandwidthTest --dtod --mode=shmoo Run the oclBandwidthTest at least 3 times (each run takes only about 2 Seconds ), there will be differences between the runs, so you get an better average value so can compare with/ without kext. (parameters gives an more intensitive internal bandwisth test , with out you get also PCI to / from gpu transferspeeds and an quick device to device(internal) bandwidth test. oclBandwidthTest.zip
  7. First you must disable (or remove those renames) all GPU related renames in the ACPI /DSDT section if you use WEG. WEG does all needed - much more clever. Complete your SMBIOS , i can only see Macmodell 18,1. Also setup your IGPU with an ig-platform ID which is connectorless (means no more an display device anymore). Even better (mire easy to handle setup) disable IGPU in BIOS , remove inject intel and use an iMacPro1,1 MacModell.
  8. mitch_de

    VirtualSMC — SMC Emulator

    It maybe because that VirtualSMC.efi is now "included /inside code " of OpenCore 0.54. So that OC Users dont need that VirtualSMC.efi driver anymore. But i may be wrong;)
  9. Yep, but if someone has an IGPU and use that .kext with dGPU he looses IGPU AGPM. Most users will have an CPU with iGPU i think. Buts no problem to add the missing IGPU part by copy it from the orig. .plist ( for used Mac Modell ) in the new .kext. PS: For my AMD RX 460 it make no difference in Speed, Temps or IDLE Temp using the kext or not. I think AMD has own AGPM in its drivers for their GPUs. So it maybe usefull for Nvidia GPU users.
  10. I tested now the generated .kext with my iMac 13,2. kext seems only add / change the GFX0 AGPM part. BUG: It deletes the IGPU AGPM part from the orig. .plist. Pic left side shows AGPM after using the kext = IGPU part is missing/deleted, rigth side before using the kext (original AppleGra..) Workaround: I copied the IGPU part (for my iMac13,2 ) into the AGPM.kext
  11. 0x11 = HEX value = 17 decimal 18 dec = 0x12
  12. For my iMac 13,2 the AppleGraphics... is same in HS und Mojave. Left of first pic is HS , right is 10.4- This App would change that into , second pic.
  13. No, your App shows error if started in HS. I buld it for 10.13 Deployment (uploaded HS Version few post before).
  14. Its only for Mojave and not High Sierra? EDIT: I build it for HS - works (generates the .kext - untested the .kext) AGPMInjector_HS.zip
  15. What do you mean with not working RGB rams? Thanks for hint with saving static light setting - i will try. I really dont like that multi colour lights show when ICUE didnt run
×