home.social

#intelarc — Public Fediverse posts

Live and recent posts from across the Fediverse tagged #intelarc, aggregated by home.social.

  1. Newest #IntelArc #GPU family member is here, the Panther Lake Arc B390... and it... purrs? 🖖 🥺 🐈‍⬛
    My OpenCL-Benchmark on the B390 measures ~7.4 TFlops FP32 and ~120GB/s memory bandwidth. hw-smi also works with the B390.
    #FluidX3D benchmarks here: github.com/ProjectPhysX/FluidX
    And the #OpenCL infos:
    -Arc B390: opencl.gpuinfo.org/displayrepo
    - Core Ultra X7 358H: opencl.gpuinfo.org/displayrepo

  2. instead of targeting gamers, these cards are aimed squarely at AI and professional workloads, signaling Intel’s strategic pivot toward high-memory, workstation-class GPUs over consumer gaming flagships.

    wccftech.com/big-battlemage-gp

    #Intel #IntelArc #Battlemage #GPU #AIHardware #WorkstationGPU #GDDR6 #GraphicsCard #TechNews #Semiconductors

  3. Intel’s long-awaited “Big Battlemage” GPU has finally arrived as the Arc Pro B70 and B65, both packing a massive 32GB of GDDR6 memory and built on the flagship BMG-G31 die, marking Intel’s most powerful discrete GPU yet.

    However, instead of targeting gamers, these cards are aimed squarely at AI and professional workloads, signaling Intel’s strategic pivot toward high-memory, workstation-class GPUs over consumer gaming flagships.

    wccftech.com/big-battlemage-gp

    #Intel #IntelArc #Battlemage #GPU #AIHardware #WorkstationGPU #GDDR6 #GraphicsCard #TechNews #Semiconductors

  4. Intel’s long-awaited “Big Battlemage” GPU has finally arrived as the Arc Pro B70 and B65, both packing a massive 32GB of GDDR6 memory and built on the flagship BMG-G31 die, marking Intel’s most powerful discrete GPU yet.
    wccftech.com/big-battlemage-gp

    #Intel #IntelArc #Battlemage #GPU #AIHardware #WorkstationGPU #GDDR6 #GraphicsCard #TechNews #Semiconductors #tech

  5. @niccolove @niccolo_ve

    Comment on this video (which us not shown on my Fedi instance, yay federation!)

    tube.kockatoo.org/w/63cGWTvhzq

    I cannot believe that that GPU is not usable for video work. That is what GPUs are made for!

    Poorly, there are a few caveats:
    - there are always more DEcoders supported than ENcoders. I guess not being able to play a video (it would play but be less efficiently decoded on the CPU) is worse than being bound to #h265 instead of #av1 for recording.
    - thus it makes sense that GPUs generally work flawlessly for gaming, as that is just displaying videos (or rendering content, which is kinda different, no idea why video codecs are more difficult than rendering methods)
    - for every video format out there exists an encoder. If you use software rendering (on the CPU), you can use ANYTHING. Like encode to modern (and free!) AV1 using #svtav1 on any old CPU.
    - if you want to encode videos on a GPU, you need drivers, but also a different encoder. On #NVIDIA you use #NVENC (NVidia ENCoder), for example nvenc_x265 for h265/HEVC videos. On #IntelArc (and afaik AMD GPUs), you use the #QSV (quick sync video) library which builds on top of #VAAPI, which is open source. So you can encode to AV1 using qsv_av1 for example.

    What encoders are supported depends on the GPU a lot. Intel Arc seems to have supported the free, efficient and future-proof AV1 codec the earliest, but also the older free codecs VP9 and VP8 (also used in webm), like in a GPU I recently bought used. My older NVIDIA GPU only supports H265, which is proprietary and often less efficient than AV1 (not always, video codecs are black magic).

    As OBS can use your GPU for video capture (encoding), the issue is purely in software support.

    I only have experience with #ffmpeg for re-encoding videos, but that is also in many video editing tools. Try if you can get them somehow.

    For NVIDIA I needed the latest proprietary drivers (#NixOS makes this extremely easy, can recommend) and ffmpeg, nvenc worked out of the box.

    For intel Arc I used the regular drivers included in the kernel, but after adding the QSV runtime to my "hardware.graphics.extraPackages" I needed to recompile the entire kernel. Using a #longterm kernel (currently 6.18) makes this more viable... longterm XD.

    #MicrosoftSurface #KDENlive #Shotcut #VideoEditing

  6. Any #Linux users on #IntelArc here?

    I bought an Intel Arc A750 and it looks fancy, works for gaming and AV1 encoding (after adding "vpl-gpu-rt" recompiling the kernel for that I guess), but I cannot control the fan at all!

    The fan mode is "automatic" and on full power all the time which is crazy loud.

    I tried everything, updated my kernel to 6.18 longterm, used a windows install to use that intel tool to upgrade the GPU firmware, no change.

    Now I plan on connecting the fan to the case fan connector and controlling it through BIOS (I can read temperature now so that is okay-ish I guess).

    I couldn't find a single report of this being a thing at all, do these GPUs really have THAT shitty drivers? I thought it was just performance issues (which I didnt really see).

    Discourse Thread:

    discourse.nixos.org/t/76041

    @[email protected] @pcmasterrace @[email protected] @buildapc @[email protected] @[email protected] @intel_arc #PcBuilding #GraphicsCard #NixOS

  7. #IntelArc asettaa todella ankaria rajoituksia sille, mitä vapaata järjestelmää kokeilukoneelle kannattaa yrittää. Kaikki #BSD:t #FreeBSD:tä myöten ovat joko kieltäytyneet pääsemästä graafiseen tilaan tai buutanneet itsensä yrittäessään; Linuxeistakin #GuixSystem'issä oli näytönpiirto-ongelmia ja lopulta #Kwin lakkasi käynnistymästä (eli työpöydälle pääsi mutta ikkunoilla ei ollut reunoja ja ne olivat vain kiinteä alue näytön vasemmassa yläkulmassa). #linux #floss #atkjuttuja

  8. I need more #JustJosh interviews like this.

    They guy just pulls out the checklist and cites it line by line to the people who are (at least somewhat) responsable.

    youtube.com/watch?v=AzGFbkKZE7A

    "No consumer is going to understand this [X and H nomenclature]. How do we solve this?"

    #CES #Technology #Computer #Intel #Computers #PC #PCHardware #CPU #APU #IntelGraphics #IntelArc #ArrowLake #PantherLake #IntelCore

  9. Intel Core Ultra Serii 3 oficjalnie. Panther Lake i proces 18A mają uratować honor „Niebieskich”

    Na ten moment czekali wszyscy inwestorzy i fani technologii. Intel podczas CES 2026 oficjalnie zaprezentował procesory Core Ultra Serii 3 (kryptonim Panther Lake).

    To pierwsze układy wyprodukowane w długo wyczekiwanym procesie technologicznym Intel 18A. Nowe czipy trafią do laptopów jeszcze w tym miesiącu.

    Pomóż nam rozwijać iMagazine – ruszyło badanie czytelnictwa 2026

    18A – być albo nie być dla Intela

    Najważniejszą informacją nie są tu same gigaherce, ale sposób produkcji. Kafelek obliczeniowy (Compute Tile) w nowych procesorach powstaje w litografii 18A. To technologia, która ma być „powrotem króla” i dowodem na to, że fabryki Intela są w stanie konkurować z tajwańskim TSMC. Sukces Panther Lake jest kluczowy nie tylko dla sprzedaży laptopów, ale dla całego planu biznesowego Intela (Intel Foundry), który chce produkować czipy dla zewnętrznych firm.

    Architektura: krok w tył, by zrobić dwa do przodu?

    Panther Lake to w pewnym sensie odwrót od filozofii znanej z poprzedniej serii (Lunar Lake / Core Ultra 200V). Tamte układy miały pamięć RAM wbudowaną w procesor (jak w Apple Silicon) i były produkowane głównie przez TSMC.

    W Serii 3 Intel wraca do korzeni, choć w nowoczesnym wydaniu. Mamy tu konstrukcję kafelkową (Foveros):

    • Compute Tile (CPU + NPU): produkcja Intel 18A (to tu dzieje się magia).
    • Platform Controller Tile (I/O): produkcja TSMC.
    • Graphics Tile (GPU): wersja mocna (12 rdzeni) – TSMC. Wersja słabsza (4 rdzenie) – starszy proces Intel 3.

     

     

    Serie „X” i podział na wydajność

    Intel wprowadza ciekawe rozróżnienie w nazewnictwie, które ma pomóc w doborze sprzętu:

    • Core Ultra X9 i X7: to układy dla ultramobilnych maszyn bez dedykowanej karty graficznej. Posiadają potężne, 12-rdzeniowe iGPU Intel Arc B390 oraz wsparcie dla superszybkiej pamięci LPDDR5x-9600. Mają jednak mniej linii PCIe (12), bo nie przewiduje się łączenia ich z zewnętrznym GPU.
    • Core Ultra 9 i 7: to układy pomyślane do laptopów gamingowych i stacji roboczych z dedykowaną grafiką (np. NVIDIA GeForce). Mają słabsze iGPU (4 rdzenie), ale za to aż 20 linii PCIe, by obsłużyć mocną kartę graficzną.

    Wydajność i bateria

    Intel składa odważne deklaracje: Panther Lake ma być do 60% szybszy w zadaniach wielowątkowych i do 77% szybszy w grafice niż poprzednik (Lunar Lake).

    Imponująco wygląda też efektywność energetyczna. Referencyjny laptop Lenovo z układem Core Ultra X9 388H pozwolił na 27 godzin streamowania Netflixa w 1080p. Oczywiście to warunki laboratoryjne, ale wynik robi wrażenie.

    W kwestii AI, zintegrowane NPU osiąga 50 TOPS. To wystarcza, by spełnić wymogi standardu Microsoft Copilot+ PC (40 TOPS), choć konkurencja ucieka nieco dalej (AMD Ryzen AI ma 60 TOPS, a nowe Snapdragony aż 80 TOPS).

    Pierwsze laptopy z Panther Lake trafią na półki sklepowe 27 stycznia.

    Masz nowego laptopa z Intel Core Ultra, a Zdjęcia Google w Chrome „tną”? Wiemy, jak to naprawić

    #CES2026 #Intel18A #IntelArc #IntelCoreUltraSeries3 #news #PantherLake #procesoryDoLaptopów
  10. Schade. Auch der aktuelle Kernel hat noch Probleme, wenn man #Firefox mit #IntelArc nutzen möchte (#LunarLake). Am Anfang läuft alles OK, aber mit der Zeit reagiert alles immer langsamer. So weit, dass selbst Tastatur-Events nicht mehr erkannt werden. Keypress ohne Release, anyone? Alle paar Sekunden geht dann mal ein Block an Events wieder durch. Sobald man Firefox beendet oder minimiert ist wieder alles normal. Selbes System läuft mit Iris Xe ohne Murren.

  11. The #IntelArc #A310 looks like the card of choice but there appears to be no card with a proper fan design available right now. There is only a super noisy #Sparkle variant with a blower fan design with several reports on this specific card to have high fan noise and a terrible fan curve. Urgh.
    So I looked at the alternative #A380 at pretty much the same price, which can be run at ~5W when idling. But there is a problem: Intel Arc cards drain 30-40W when there is no #ASMP enabled to allow the card to low power states: intel.com/content/www/us/en/su
    Well, ASMP is only supported on newer hardware so I had to to check my mainboard (AMD B520 chip) for support. (2/?)

  12. The #IntelArc #A310 looks like the card of choice but there appears to be no card with a proper fan design available right now. There is only a super noisy #Sparkle variant with a blower fan design with several reports on this specific card to have high fan noise and a terrible fan curve. Urgh.
    So I looked at the alternative #A380 at pretty much the same price, which can be run at ~5W when idling. But there is a problem: Intel Arc cards drain 30-40W when there is no #ASMP enabled to allow the card to low power states: intel.com/content/www/us/en/su
    Well, ASMP is only supported on newer hardware so I had to to check my mainboard (AMD B520 chip) for support. (2/?)

  13. The #IntelArc #A310 looks like the card of choice but there appears to be no card with a proper fan design available right now. There is only a super noisy #Sparkle variant with a blower fan design with several reports on this specific card to have high fan noise and a terrible fan curve. Urgh.
    So I looked at the alternative #A380 at pretty much the same price, which can be run at ~5W when idling. But there is a problem: Intel Arc cards drain 30-40W when there is no #ASMP enabled to allow the card to low power states: intel.com/content/www/us/en/su
    Well, ASMP is only supported on newer hardware so I had to to check my mainboard (AMD B520 chip) for support. (2/?)

  14. The #IntelArc #A310 looks like the card of choice but there appears to be no card with a proper fan design available right now. There is only a super noisy #Sparkle variant with a blower fan design with several reports on this specific card to have high fan noise and a terrible fan curve. Urgh.
    So I looked at the alternative #A380 at pretty much the same price, which can be run at ~5W when idling. But there is a problem: Intel Arc cards drain 30-40W when there is no #ASMP enabled to allow the card to low power states: intel.com/content/www/us/en/su
    Well, ASMP is only supported on newer hardware so I had to to check my mainboard (AMD B520 chip) for support. (2/?)

  15. The #IntelArc #A310 looks like the card of choice but there appears to be no card with a proper fan design available right now. There is only a super noisy #Sparkle variant with a blower fan design with several reports on this specific card to have high fan noise and a terrible fan curve. Urgh.
    So I looked at the alternative #A380 at pretty much the same price, which can be run at ~5W when idling. But there is a problem: Intel Arc cards drain 30-40W when there is no #ASMP enabled to allow the card to low power states: intel.com/content/www/us/en/su
    Well, ASMP is only supported on newer hardware so I had to to check my mainboard (AMD B520 chip) for support. (2/?)

  16. In order of recommendation:

    - $1,200+ Laptop with an RTX
    - $700 Legion Go / $800 ROG Ally X
    - $500 PlayStation 5
    - $400 Steam Deck
    - $300 Xbox Series S

    That's is for gaming appliances in 2025. Hopefully things change for the better Q4 2025, but no #PCGaming until then.

    #Videogames #Gaming #Games #Laptop #Laptops #RTX #NVIDIA #RX9070 #Radeon #GeForce #AMD #LegionGo #ROGAlly #SteamDeck #IntelArc #B570 #PlayStation5 #PS5 #Console #Consoles #XboxSeriesS #XboxSeries #Xbox #PC #PCHardware #PCGames

  17. «Stop buying NVIDIA and their overpriced cards!»

    Unless AMD disrupts the market like Intel is doing it with the #B570, I expect another year with this list being all NVIDIA.

    #Technology #AMD #NVIDIA #GPU #dGPU #Computer #Computers #PC #PCHardware #Hardware #Videogames #Gaming #Games #Steam #Valve #RTX #RTX3060 #RTX5080 #RTX5090 #GeForce #Radeon #IntelArc #Arc #ArcB570

  18. Bah. Looks like have hit the limits of the P400’s NVENC transcoder; hit rates below 1.00x for first time... it's put in a few years.

    The P400 RAM usage is about 1 gb with 4 streams, but 100% on encode/decode. The P400 has 1 Pascal (6th gen) nvenc block and 2gb RAM.

    Have an Arc A380 coming in now.

    A380 should be able to handle quite a bit more, has two MFX engines (quicksync) & 6gb RAM.

    Time will tell.

    [ #P400 #A380 #nvenc #qsv #IntelArc #transcoding ]

  19. Another great news for this 2025: #ZLUDA lives

    I'm eager for this library to take off. While NVIDIA dominates the market, there are decent iGPU like the Radeon 680M/780M/870M and Intel Arc on Ultra 200 Series.

    phoronix.com/news/ZLUDA-v4-Rel

    #CUDA #NVIDIA #RTX #AI #ML #LM #LLM #LargeLanguageModels #AMD #Radeon #Intel #IntelArc #Battlemage #B570 #PC #PCHardware #Hardware #GPU #iGPU #dGPU

  20. 從早上就在補各大媒體對於 #intel ARC #B580 的評測
    先說結論:intel 你出了張好卡

    intel Arc B580 本身的效能並不出色
    但在臺幣 7000 - 10000 區間
    新架構、新技術以及大容量的顯示記憶體
    #AMD#nvidia 已無心經營的價格區段
    intel 這次切入的市場定位著實漂亮

    2022 年 intel A770 / A750 一樣切入臺幣一萬區間的市場
    卻沒有辦法獲得市場的讚揚
    Allen 看來最主要還是在於驅動程式的不純熟以及首次上市的過度溢價
    最傷的還是系列產品上市的遲滯
    在另一篇小聊 #A580 的文中
    Allen 就提到 A580 的上市價格與當時的 A750 上市市價相當
    就使得定位發生錯位與模糊

    回到這次 B580 的上市
    兩年過去 intel 的驅動程式已經打磨許久
    另外兩家在該區段市場的忽視
    以及抓準了聖誕假期的時間上市
    都讓這次 B580 佔盡了風頭

    目前臺灣也已經上市
    價格落在 8000 - 10000 區間
    晚點再來做同價位卡的橫向比較

    #Allen講看看 #IntelArc #GPU #顯示卡

  21. I wonder if the new Intel Arc #Battlemage GPUs might bring me to retire my Nvidia 1080ti. I have to wait for benchmarks. And I have to find out how well it is supported on Linux. I hope this generation is good. I'd really like some competition in the GPU space.

    #Intel #IntelArc #GPU #B580 #B570

  22. Honestly, I don't see the point of having a third-party #SteamDeck going for more than $600. Makes no freaking sense.

    After that price these compete in the #laptop space, where you can find superior SoC on all sides: #Zen4, #Zen3, #MeteorLake or #LunarLake. These win hands down no contest.

    #Hardware #Console #AMD #Consoles #iGPU #Radeon #Intel #PCHardware #Steam #RogAlly #LenovoLegionGo #LegionGo #IntelArc #Videogames #Gaming #Games #PCGaming #PCGames

  23. Honestly, I don't see the point of having a third-party #SteamDeck going for more than $600. Makes no freaking sense.

    After that price these compete in the #laptop space, where you can find superior SoC on all sides: #Zen4, #Zen3, #MeteorLake or #LunarLake. These win hands down no contest.

    #Hardware #Console #AMD #Consoles #iGPU #Radeon #Intel #PCHardware #Steam #RogAlly #LenovoLegionGo #LegionGo #IntelArc #Videogames #Gaming #Games #PCGaming #PCGames

  24. Honestly, I don't see the point of having a third-party #SteamDeck going for more than $600. Makes no freaking sense.

    After that price these compete in the #laptop space, where you can find superior SoC on all sides: #Zen4, #Zen3, #MeteorLake or #LunarLake. These win hands down no contest.

    #Hardware #Console #AMD #Consoles #iGPU #Radeon #Intel #PCHardware #Steam #RogAlly #LenovoLegionGo #LegionGo #IntelArc #Videogames #Gaming #Games #PCGaming #PCGames

  25. Honestly, I don't see the point of having a third-party #SteamDeck going for more than $600. Makes no freaking sense.

    After that price these compete in the #laptop space, where you can find superior SoC on all sides: #Zen4, #Zen3, #MeteorLake or #LunarLake. These win hands down no contest.

    #Hardware #Console #AMD #Consoles #iGPU #Radeon #Intel #PCHardware #Steam #RogAlly #LenovoLegionGo #LegionGo #IntelArc #Videogames #Gaming #Games #PCGaming #PCGames

  26. Honestly, I don't see the point of having a third-party #SteamDeck going for more than $600. Makes no freaking sense.

    After that price these compete in the #laptop space, where you can find superior SoC on all sides: #Zen4, #Zen3, #MeteorLake or #LunarLake. These win hands down no contest.

    #Hardware #Console #AMD #Consoles #iGPU #Radeon #Intel #PCHardware #Steam #RogAlly #LenovoLegionGo #LegionGo #IntelArc #Videogames #Gaming #Games #PCGaming #PCGames

  27. @python I measured peak ~1.2GLUPs/s with #FP16​S memory compression, 67% efficient regarding 136GB/s RAM bandwidth (8533 MT/s). That makes #Intel Lunar Lake 140V 1.7x faster than the Meteor Lake 185H iGPU. It's about on par with #IntelArc A380, RX 6500 XT, GTX 1050M Ti. Very cool to see an iGPU finally be competitive with entry level discrete #GPU​s!
    github.com/ProjectPhysX/FluidX

  28. So, final update on #IntelArc A380 in #Kubevirt/#Kubernetes.

    To put it lightly, the whole thing was a bust. Likely at least partially due to my motherboard (whatever comes with a Dell R730) not supporting resizable bar. To add to that, while Linux will try its best to resize the PCIe BAR even if the BIOS feature isn't enabled, it seems the architecture of the Intel Arc cards blocks this (something to do with the card having a built-in PCIe switch that has its own resizable bar which blocks the actual GPU's bar from being resized... or something?

    Everything I've seen online seems to be point to resizable bar being almost a hard requirement for the Arc cards to do any good. So it's perhaps due to not being able to do that that the performance of the card has been absolutely terrible.

    #DavinciResolve Studio would barely even run in a Linux VM (which may be as much to do with the vendor than the GPU itself). On Windows the performance was average at best, and the GPU wasn't able to handle rendering to any of the intermediate codecs (like DNxHR or Cineform) without Davinci crashing due to running out of VRAM.

    I have parts coming somewhat soon to build a nice tidy 1U server based around the #TopTon #7840HS board, which looks exciting. I'm tempted to pop the GPU in there, but at the same time I'm pretty sure the #780M iGPU actually has better performance than the A380, so I'm probably just going to utilise the iGPU for that.

    So might be sending the A380 onto its next (and hopefully Resizable Bar capable) home shortly.