PC Ekspert Forum

PC Ekspert Forum (https://forum.pcekspert.com/index.php)
-   DIY i Tinkering (https://forum.pcekspert.com/forumdisplay.php?f=74)
-   -   ARM (https://forum.pcekspert.com/showthread.php?t=306559)

Bubba 13.07.2024. 23:32

SolidRun MACCHIATObin Single Shot
 
SolidRun MACCHIATObin Single Shot


Evo jos jedan kasnonocni ARM projektic - SolidRun MACCHIATObin Single Shot sa Marvell 88F8040@2GHz, 16GB DDR4 ECC RAM-om i malo SSD storagea. Moram vidjeti hoce li biti raspolozen za M2 na ovom PCI-e konektoru, pa da malo prosirim lab s jos ARM igracaka...

CPU nije neki spektakl, ali mi je ta ARMADA dosta poznata pa se i danas moze izvuci sasvim respektabilne performanse za malo vata...

calypso 14.07.2024. 14:47

De ti meni reci Bubba, cemu sluze te stvarcice koje slazes? Jel imaju neku namjenu ili samo slazes da bi slagao? :)

Bubba 14.07.2024. 21:21

Citiraj:

Autor calypso (Post 3752119)
De ti meni reci Bubba, cemu sluze te stvarcice koje slazes? Jel imaju neku namjenu ili samo slazes da bi slagao? :)

Nemam vise mladosti u sebi za hardverski larpurlartizam, tako da ovo ima usecase. :)

ARM workstation sluzi kao dev i build masina, s obzirom da ipak ima dosta soka u sebi (primarno memorije, CPU vise manje), tako da sto se tice razvoja, sluzi kao nativni stroj za mainlineing ARM specificnih funkcionalnosti u jedan od top 10 TIOBE jezika, dok za build pricu upravo ide implementacija CI/CD procesa kako bi se ofriskali tradicionalno zastarjeli ARM paketi za popularne distribucije (trenutno ciljamo na deb pakete).

Takodjer, u produkciji je iEi PUZZLE-M801 (ima istu Armadu kao i ove ITX ploce gore) na kojemu je Postgres, NATS i jos par servisa gdje testiram kako se ponasaju u divljini.

Ove dvije plocice ce za sada biti u R&D segmentu, jer bih primarno volio isprobati na njima kako ce se ponasati neki oblik virtualizacije, a temeljem iskustava do sada, u full loadu mogu deliverati zadovoljavajuce performanse kao x86 masine koje su u idleu.

Bubba 20.07.2024. 00:29

MacchiatoBin Workstation


Geekbench6: https://browser.geekbench.com/v6/cpu/6980334

Code:

arm-averell
    description: Computer
    product: Armada 8040 MacchiatoBin
    vendor: SolidRun
    version: Rev. 1.3
    serial: Serial Not Set
    width: 64 bits
    capabilities: smbios-3.2.0 dmi-3.2.0 smp cp15_barrier setend swp tagged_addr_disabled
    configuration: boot=normal uuid=97c93925-1273-4f03-9f75-2f2b7ed19480
  *-core
      description: Motherboard
      physical id: 0
    *-firmware
          description: BIOS
          vendor: EDK2
          physical id: 3
          version: EDK2 SR 1.0
          date: Jul 19 2024
          size: 96KiB
          capabilities: pci pnp upgrade bootselect acpi usb uefi
    *-board
          description: Motherboard
          product: Armada 8040 MacchiatoBin
          vendor: SolidRun
          physical id: 1004
          version: Rev. 1.3
          serial: Serial Not Set
          slot: Base of Chassis
        *-cpu
            description: CPU
            product: ARM (0xd08)
            vendor: Marvell
            physical id: 1006
            bus info: cpu@0
            version: Cortex-A72
            slot: Socket type unknown
            size: 2GHz
            capacity: 2200MHz
            capabilities: lm
            configuration: cores=4 enabledcores=4 threads=4
          *-cache:0
                description: L1 cache
                physical id: 1000
                slot: L1 Instruction
                size: 48KiB
                capacity: 48KiB
                capabilities: internal instruction
                configuration: level=1
          *-cache:1
                description: L2 cache
                physical id: 1002
                slot: L2
                size: 512KiB
                capacity: 512KiB
                capabilities: internal write-back unified
                configuration: level=2
          *-cache:2
                description: L3 cache
                physical id: 1003
                slot: L3
                size: 1MiB
                capacity: 1MiB
                capabilities: internal write-back unified
                configuration: level=3
    *-cache
          description: L1 cache
          physical id: 1001
          slot: L1 Data
          size: 32KiB
          capacity: 32KiB
          capabilities: internal data
          configuration: level=1
    *-memory
          description: System Memory
          physical id: 1007
          slot: System board or motherboard
          size: 15GiB
        *-bank
            description: Row of chips DDR4 1200 MHz (0.8 ns)
            physical id: 0
            slot: DIMM SLOT
            size: 15GiB
            width: 32 bits
            clock: 1200MHz (0.8ns)
    *-display
          description: VGA compatible controller
          product: RV730 PRO [Radeon HD 4650]
          vendor: Advanced Micro Devices, Inc. [AMD/ATI]
          physical id: 0
          bus info: pci@0000:00:00.0
          logical name: /dev/fb0
          version: 00
          width: 64 bits
          clock: 33MHz
          capabilities: pm pciexpress msi vga_controller bus_master cap_list rom fb
          configuration: depth=32 driver=radeon latency=0 mode=1920x1080 resolution=1920,1080 visual=truecolor xres=1920 yres=1080
          resources: iomemory:80-7f iomemory:80-7f irq:54 memory:800000000-80fffffff memory:810000000-81000ffff ioport:1000(size=256) memory:c0000000-c001ffff
    *-multimedia
          description: Audio device
          product: RV710/730 HDMI Audio [Radeon HD 4000 series]
          vendor: Advanced Micro Devices, Inc. [AMD/ATI]
          physical id: 0.1
          bus info: pci@0000:00:00.1
          logical name: card0
          logical name: /dev/snd/controlC0
          logical name: /dev/snd/hwC0D0
          logical name: /dev/snd/pcmC0D3p
          version: 00
          width: 64 bits
          clock: 33MHz
          capabilities: pm pciexpress msi bus_master cap_list
          configuration: driver=snd_hda_intel latency=0
          resources: iomemory:80-7f irq:55 memory:810010000-810013fff
        *-input
            product: HDA ATI HDMI HDMI/DP,pcm=3
            physical id: 0
            logical name: input1
            logical name: /dev/input/event1
    *-pnp00:00
          product: PnP device PNP0c02
          physical id: 1
          capabilities: pnp
          configuration: driver=system
    *-scsi
          physical id: 2
          logical name: scsi2
          capabilities: emulated
        *-disk
            description: ATA Disk
            product: WDC WDS500G2B0A
            vendor: Western Digital
            physical id: 0.0.0
            bus info: scsi@2:0.0.0
            logical name: /dev/sda
            version: 90WD
            serial: 20101S465005
            size: 465GiB (500GB)
            capabilities: removable
            configuration: ansiversion=5 logicalsectorsize=512 sectorsize=512
          *-medium
                physical id: 0
                logical name: /dev/sda
                size: 465GiB (500GB)
                capabilities: gpt-1.00 partitioned partitioned:gpt
                configuration: guid=b02ed44d-0087-4e3d-a8ea-826db3243d77
              *-volume:0 UNCLAIMED
                  description: Windows FAT volume
                  vendor: mkfs.fat
                  physical id: 1
                  version: FAT32
                  serial: 072c-7dea
                  size: 1073MiB
                  capacity: 1074MiB
                  capabilities: boot fat initialized
                  configuration: FATs=2 filesystem=fat
              *-volume:1
                  description: EXT4 volume
                  vendor: Linux
                  physical id: 2
                  logical name: /dev/sda2
                  logical name: /
                  logical name: /var/snap/firefox/common/host-hunspell
                  version: 1.0
                  serial: 8eff69d7-2632-4c81-ba4e-4cccd09ab370
                  size: 464GiB
                  capabilities: journaled extended_attributes large_files huge_files dir_nlink recover 64bit extents ext4 ext2 initialized
                  configuration: created=2024-07-19 18:29:09 filesystem=ext4 lastmountpoint=/ modified=2024-07-19 19:09:09 mount.fstype=ext4 mount.options=ro,noexec,noatime mounted=2024-07-19 19:09:11 state=mounted
  *-usbhost:0
      product: xHCI Host Controller
      vendor: Linux 5.15.0-116-generic xhci-hcd
      physical id: 1
      bus info: usb@1
      logical name: usb1
      version: 5.15
      capabilities: usb-2.00
      configuration: driver=hub slots=1 speed=480Mbit/s
  *-usbhost:1
      product: xHCI Host Controller
      vendor: Linux 5.15.0-116-generic xhci-hcd
      physical id: 2
      bus info: usb@2
      logical name: usb2
      version: 5.15
      capabilities: usb-3.00
      configuration: driver=hub slots=1 speed=5000Mbit/s
  *-usbhost:2
      product: xHCI Host Controller
      vendor: Linux 5.15.0-116-generic xhci-hcd
      physical id: 3
      bus info: usb@3
      logical name: usb3
      version: 5.15
      capabilities: usb-2.00
      configuration: driver=hub slots=1 speed=480Mbit/s
  *-usbhost:3
      product: xHCI Host Controller
      vendor: Linux 5.15.0-116-generic xhci-hcd
      physical id: 4
      bus info: usb@4
      logical name: usb4
      version: 5.15
      capabilities: usb-3.00
      configuration: driver=hub slots=1 speed=5000Mbit/s
  *-usbhost:4
      product: xHCI Host Controller
      vendor: Linux 5.15.0-116-generic xhci-hcd
      physical id: 5
      bus info: usb@5
      logical name: usb5
      version: 5.15
      capabilities: usb-2.00
      configuration: driver=hub slots=1 speed=480Mbit/s
    *-usb
          description: USB hub
          product: USB 2.0 Hub
          vendor: Terminus Technology Inc.
          physical id: 1
          bus info: usb@5:1
          version: 1.11
          capabilities: usb-2.00
          configuration: driver=hub maxpower=100mA slots=4 speed=480Mbit/s
        *-usb:0
            description: Mass storage device
            product: DataTraveler 3.0
            vendor: Kingston
            physical id: 2
            bus info: usb@5:1.2
            logical name: scsi4
            version: 1.10
            serial: E0D55E6B6472E4C109141306
            capabilities: usb-2.10 scsi emulated scsi-host
            configuration: driver=usb-storage maxpower=300mA speed=480Mbit/s
          *-disk
                description: SCSI Disk
                product: DataTraveler 3.0
                vendor: Kingston
                physical id: 0.0.0
                bus info: scsi@4:0.0.0
                logical name: /dev/sdb
                version: PMAP
                serial: EE04DC5150E9
                size: 115GiB (124GB)
                capabilities: removable
                configuration: ansiversion=6 logicalsectorsize=512 sectorsize=512
              *-medium
                  physical id: 0
                  logical name: /dev/sdb
                  size: 115GiB (124GB)
                  capabilities: partitioned partitioned:dos
                  configuration: signature=a057aa1e
                *-volume:0
                      description: HPFS/NTFS partition
                      physical id: 1
                      logical name: /dev/sdb1
                      logical name: /media/bubba/Ventoy
                      capacity: 115GiB
                      capabilities: primary bootable
                      configuration: mount.fstype=exfat mount.options=rw,nosuid,nodev,relatime,uid=1000,gid=1000,fmask=0022,dmask=0022,iocharset=utf8,errors=remount-ro state=mounted
                *-volume:1
                      description: Windows FAT volume
                      vendor: mkfs.fat
                      physical id: 2
                      logical name: /dev/sdb2
                      version: FAT16
                      serial: 3f32-27f5
                      size: 31MiB
                      capacity: 32MiB
                      capabilities: primary boot fat initialized
                      configuration: FATs=2 filesystem=fat label=VTOYEFI
        *-usb:1
            description: Keyboard
            product: Lenovo Lenovo Traditional USB Keyboard
            vendor: Lenovo
            physical id: 3
            bus info: usb@5:1.3
            logical name: input0
            logical name: /dev/input/event0
            logical name: input0::capslock
            logical name: input0::numlock
            logical name: input0::scrolllock
            version: 1.00
            capabilities: usb-2.00 usb
            configuration: driver=usbhid maxpower=100mA speed=2Mbit/s
        *-usb:2
            description: Mouse
            product: USB OPTICAL MOUSE
            physical id: 4
            bus info: usb@5:1.4
            logical name: input2
            logical name: /dev/input/event2
            logical name: /dev/input/mouse0
            version: 1.00
            capabilities: usb-1.10 usb
            configuration: driver=usbhid maxpower=100mA speed=2Mbit/s
  *-usbhost:5
      product: xHCI Host Controller
      vendor: Linux 5.15.0-116-generic xhci-hcd
      physical id: 6
      bus info: usb@6
      logical name: usb6
      version: 5.15
      capabilities: usb-3.00
      configuration: driver=hub slots=1 speed=5000Mbit/s
  *-mmc0
      description: MMC Host
      physical id: 7
      logical name: mmc0
    *-device
          description: SD/MMC Device
          product: 8GTF4R
          vendor: Unknown (21)
          physical id: 1
          bus info: mmc@0:0001
          date: 03/2018
          serial: 996524155
          capabilities: mmc
        *-interface:0
            physical id: 1
            logical name: /dev/mmcblk0rpmb
        *-interface:1
            physical id: 2
            logical name: /dev/mmcblk0
            size: 7818182656
            configuration: logicalsectorsize=512 sectorsize=512
  *-mmc1
      description: MMC Host
      physical id: 8
      logical name: mmc1
    *-device
          description: SD Card
          product: SS08G
          vendor: SanDisk
          physical id: 59b4
          logical name: /dev/mmcblk1
          version: 8.0
          date: 06/2015
          serial: 1080590894
          size: 7583MiB (7951MB)
          capabilities: sd
          configuration: logicalsectorsize=512 sectorsize=512
  *-network:0 DISABLED
      description: Ethernet interface
      physical id: 9
      logical name: enamrvl110i0
      serial: 4a:cb:f6:a0:7f:f8
      capacity: 10Gbit/s
      capabilities: ethernet physical tp aui bnc mii fibre 10000bt-fd autonegotiation
      configuration: autonegotiation=off broadcast=yes driver=mvpp2 driverversion=1.0 duplex=half link=no multicast=yes port=MII
  *-network:1 DISABLED
      description: Ethernet interface
      physical id: a
      logical name: enamrvl110i1
      serial: 7e:69:d2:6e:fc:1f
      capacity: 10Gbit/s
      capabilities: ethernet physical tp aui bnc mii fibre 10000bt-fd autonegotiation
      configuration: autonegotiation=off broadcast=yes driver=mvpp2 driverversion=1.0 duplex=half link=no multicast=yes port=MII
  *-network:2
      description: Ethernet interface
      physical id: b
      logical name: eth2
      serial: 7e:69:d2:6e:fc:1f
      size: 1Gbit/s
      capacity: 1Gbit/s
      capabilities: ethernet physical tp mii fibre 10bt 10bt-fd 100bt 100bt-fd 1000bt-fd autonegotiation
      configuration: autonegotiation=on broadcast=yes driver=mvpp2 driverversion=1.0 duplex=full ip=172.16.222.111 link=yes multicast=yes port=twisted pair speed=1Gbit/s
  *-network:3 DISABLED
      description: Ethernet interface
      physical id: c
      logical name: eth3
      serial: 7e:69:d2:6e:fc:1f
      capacity: 2500Mbit/s
      capabilities: ethernet physical mii 2500bt-fd autonegotiation
      configuration: autonegotiation=on broadcast=yes driver=mvpp2 driverversion=1.0 link=no multicast=yes port=MII


The Exiled 28.08.2024. 18:15

Citiraj:

AmpereOne A192-32X Benchmarks: 192 Core ARM Server Performance & Power Efficiency:kafa:
Citiraj:

AmpereOne reviewer system is based on a Supermicro ARS-211M-NR R13SPD that is an air-cooled 2U server solution for AmpereOne processors -- capable of handling up to the top-end 400 Watt processors. The Supermicro ARS-211M-NR R13SPD can handle up to four double-width GPUs, 16 DIMM slots in total for up to 2TB of DDR5-5200 memory with one DIMM per channel or up to 4TB of RAM at two DIMMs per channel but reduced then to DDR5-4400 speeds. The AmpereOne server can boot standard AArch64 Linux distributions like Ubuntu 24.04 LTS and Fedora Server 40. For the initial benchmarking I was using Ubuntu 24.04 LTS. This review server shipped with 8 x 64GB DDR5-5200 DIMMs, Samsung NVMe MZQL23T8HCLS-00A07 + MZ1L2960HCJR-00A07 storage, and using the top-end AmpereOne A192-32X processor. The AmpereOne A192-32X boasts 192 AmpereOne cores, a 3.2GHz clock frequency, a rated 276 Watt usage power, and carries a $5,555 USD suggested price.

The AmpereOne A192-32X performed admirably in matching the performance of a dual Intel Xeon 6766E Sierra Forest configuration overall but behind the Xeon 6780E 2P configuration. The AmpereOne A192-32X in a single socket configuration was also behind the single EPYC 9654 Genoa / 9754 Bergamo / 9684X Genoa-X processors. Going from the Ampere Altra Max M128-30 to AmpereOne A192-32X was 1.56x the performance overall. That's with 1.5x the core count in going from Ampere Altra Max 128 cores to 192 cores with the current top-end AmpereOne SKUs. As shown in the few single-threaded and lightly threaded workloads, there wasn't too much IPC uplift going to AmpereOne from Ampere Altra cores. With AmpereOne is also the move to DDR5 memory from DDR4 and other improvements.

When looking at the CPU power consumption over the span of all the benchmarks carried out, the AmpereOne A192-32X on average was consuming 230 Watts with a peak of 401 Watts, up from a 166 Watt average for the Ampere Altra Max M128-30 with a 166 Watt average and 252 Watt peak. The most significant difference with AmpereOne was unfortunately the much higher CPU power consumption at idle and lightly threaded scenarios. The AmpereOne A192-32X never consumed less than 101 Watts even in the brief periods of being idle... Meanwhile the single Intel / AMD server processors could idle in the 10~20'ish Watt range. The Ampere Altra Max M128-30 could sip as little as 21 Watts. But the AmpereOne didn't drop below 100 Watts. The EPYC 9754 Bergamo had a similar power consumption overall to the AmpereOne A192-32X with a 225 Watt average and 397 Watt peak. The EPYC 9754 would consume just 10 Watts at idle.

Overall this was a better than expected first look at AmpereOne. It's great finally seeing AmpereOne hands-on and the performance being competitive. If AmpereOne had ramped up last year well before Intel Sierra Forest it would have looked extremely compelling with its 192 core part and launching to a very different landscape. Now it's appearing with Intel Sierra Forest having been "launched" since earlier this summer, Intel Granite Rapids coming up soon, and AMD EPYC Turin also due out in the coming months. It will be interesting to see if Ampere Computing can achieve robust availability on AmpereOne processors and platforms before Granite Rapids and EPYC Turin. Hopefully firmware updates can also help with AmpereOne reducing its idle CPU power usage.
Izvor: Phoronix

Sinac 23.10.2024. 08:59

Ima li itko tu na forumu Windows on ARM upogonjen? Čisto me zanimaju korisnička iskustva iz prve ruke. Vidim da je Lenovo zagazio u te vode s poslovnom serijom laptopa, prvo X13s i s nešto novijim T14s Gen6.
Nadam se da će se stvoriti dobra baza korisnika jer u ovom trenu mi situacija sliči na onu s Windows Mobile uređajima, sjajna hardverska podloga s premalim brojem aplikacija koju je Microsoft sa svakom novom verzijom OS-a smanjivao jer nije dozvoljavao nadogradnje starijih uređaja. Ne bi bilo dobro da se stvar ugasi kao Windows RT.

The Exiled 23.10.2024. 09:44

Izgleda da su sami sebi smjestili, kaj po pitanju ionak sfušanih Windows ARM peripetija, kaj po pitanju zakulisnih nadmudrivanja, mada je hardver solidan.
Citiraj:

  • Qualcomm abruptly cancels Snapdragon X Elite dev kit — refunds customers, ends sales and support for the device immediately

  • Legal dispute after Nuvia takeover: Arm withdraws Qualcomm's architectural license
    Citiraj:

    The background to the current development is the takeover of Nuvia by Qualcomm and the further development of the Phoenix CPU core, which was developed on the basis of an architecture license acquired from Nuvia and was originally intended to be used in the server segment, but is used by Qualcomm in an adapted form for a notebook design becomes. Qualcomm believes that by acquiring Nuvia, it has also taken over the architectural license and has the right to adapt the design developed by Nuvia to its own requirements.

    Arm, in turn, argues that Nuvia had an architectural license specifically for server designs and a transfer to Qualcomm should not have happened without Arm's prior approval. A new license agreement would probably have been linked to this, as Qualcomm is currently using the further developed Phoenix design in the mass market for notebooks instead of servers and has recently completed and announced the second Oryon generation for a smartphone chip based on the architecture license. Nuvia's former architectural licenses were terminated back in February 2023 after Arm and Qualcomm were unable to reach an agreement.
    Izvor: ComputerBase


Sinac 23.10.2024. 10:17

Citiraj:

Autor The Exiled (Post 3769811)
Izgleda da su sami sebi smjestili, kaj po pitanju ionak sfušanih Windows ARM peripetija, kaj po pitanju zakulisnih nadmudrivanja, mada je hardver solidan.

Ne znam što uopće reći na ovo. Nadam se da iz svega neće ispasti šuplja priča i da će Microsoft nastaviti s razvojem WOA. Bila bi šteta da se na kraju na svim tim računalima sa Snapdragonom vrti samo Linux.
Ne kužim zašto Arm i Nuvia/Qualcomm nisu uspjeli riješiti licencu i za notebook/desktop segment, ali to je njihova stvar.

Inače, ovo lijepo izgleda:
https://windowsonarm.org/

Bubba 23.10.2024. 18:12

Citiraj:

Autor Sinac (Post 3769800)
Ima li itko tu na forumu Windows on ARM upogonjen?

Ja sam se neki dan igrao sa Surface Proom nekoliko sati.

Microsoft Surface Pro - ARM


S obzirom da znam kakav je Win API kod iza ovoga, ugodno sam iznenadjen da sam ga pokrenuo bez ijednog pitanja i s punom funkcionalnoscu (zaboravio sam okinuti fotku nakon updatea).

Jedini razlog zasto nisam kupio uredjaj je zato sto podrzava samo Windowse. :D Ali generalni look&feel je bio zapravo sjajan, moram priznati. Baterija onako kako bi covjek ocekivao od M serije Macova.

Sinac 23.10.2024. 19:56

E to me zanimalo, pogotovo ovo oko baterije, hvala.

Nadam se da će se Qualcomm izvući iz ove zavrzlame s licencama i da će WOA zaživjeti kako spada. Nekako mi se čini da bi takvi SoC laptopi mogli biti pristojna računala sa zavidnim performansama.

Bubba 23.10.2024. 21:52

Citiraj:

Autor Sinac (Post 3769908)
E to me zanimalo, pogotovo ovo oko baterije, hvala.

Uzmi to sa zrnom soli jer samo doslovce samo surfao i kliketao po Windowsima, pa treba malo bolje istraziti sto se tu zapravo dogadja.

Ali poucen iskustvom Chromebooka, za kojeg zaboravim da mu treba punjac - autonomija je ocekivano sjajna.

Citiraj:

Nadam se da će se Qualcomm izvući iz ove zavrzlame s licencama i da će WOA zaživjeti kako spada. Nekako mi se čini da bi takvi SoC laptopi mogli biti pristojna računala sa zavidnim performansama.
ARM bi mogao prije ili kasnije zavrsiti kao Rambus sa tim svojim IP smicalicama.

Pogotovo kada se RISC-V lagano probija i gotovo da ni ne kaska hardverom koliko kronicno softverom za raznim ARM iteracijama. Uskoro ce na PcE (nadam se da cu uhvatiti vremena za vikend) osvanuti jedan sareni RISC-V roundup sa Mango-Pi RBP Zero formata, DeepComputing DC-ROMA laptopa i SiPeedovog LicheePi4A SBC-a s cak 16GB RAM-a. I sâm se veselim vidjeti neke rezultate.

PS, da se osvrnem na ovo:

Citiraj:

Bila bi šteta da se na kraju na svim tim računalima sa Snapdragonom vrti samo Linux.
U principu, ako pod "svim tim racunalima" ne mislis na Android, onda se zapravo jos uvijek niti na jednom, IMNSHO, nazalost, ne vrti Linux uopce. :D

Sinac 24.10.2024. 08:09

Citiraj:

Autor Bubba (Post 3769926)
U principu, ako pod "svim tim racunalima" ne mislis na Android, onda se zapravo jos uvijek niti na jednom, IMNSHO, nazalost, ne vrti Linux uopce. :D

Raspberry Pi vrti Linux bez problema. Mislio sam općenito na ARM platformu koju netko želi iskoristiti kao zamjenu za klasično Windows računalo.
Ova priča s WOA mi se čini s hardverske strane dobar smjer u kojem je Qualcomm krenuo jer malo podsjeća na Appleov zaokret koji je nastao s M1/2/3 (i uskoro M4) procesorom. Solidna brzina rada, manje zagrijavanje i manja potrošnja el. energije, a baterija omogućuje zavidnu autonomiju kod prijenosnih računala.

Bubba 24.10.2024. 10:42

Citiraj:

Autor Sinac (Post 3769962)
Raspberry Pi vrti Linux bez problema.

Vrti ga i 90% mreznih uiredjaja, ali to nam nije tema.

Citiraj:

Mislio sam općenito na ARM platformu koju netko želi iskoristiti kao zamjenu za klasično Windows računalo.
Ali RbP je tu besmislen jer mozda tek s 5icomm nesto suvislo mozes raditi. Uostalom, cijena s dovoljno RAM-a je previsoka jer mozes uzeti rabljeno racunalo daleko vecih performansi, prosirivosti i vjerojatno zanemarive razlike u velicini za vecinu korisnickih slucajeva.

Citiraj:

Ova priča s WOA mi se čini s hardverske strane dobar smjer u kojem je Qualcomm krenuo jer malo podsjeća na Appleov zaokret koji je nastao s M1/2/3 (i uskoro M4) procesorom. Solidna brzina rada, manje zagrijavanje i manja potrošnja el. energije, a baterija omogućuje zavidnu autonomiju kod prijenosnih računala.
Upravo to. Ali kako rekoh, ja se svesrdno nadam Linux podrsci. <3 :)

The Exiled 01.11.2024. 15:49

Citiraj:

nVidia could enter the AI PC market in 2025 with its own Arm-based CPU:kafa:
Citiraj:

Reports indicating that nVidia plans to compete with Intel, Qualcomm, and AMD in the AI PC processor market have circulated for about a year. As Arm's spat with Qualcomm intensifies and the latter's Arm PC exclusivity license nears its end, other companies, including MediaTek and nVidia, are expected to broaden the sector next year. DigiTimes reports that nVidia's long-rumored AI PC chip will emerge in the third quarter of 2025. The company, which has become one of the most valuable on Earth due to surging demand for its AI graphics processors, is reportedly developing an Arm-based platform for consumer PCs.

According to sources, the high-end platform will incorporate CPUs and GPUs. The report vaguely states that a consumer platform will debut in September 2025 and a commercial platform will follow in March 2026. This might refer to a platform for end users followed by a lineup of enterprise products or indicate a late 2025 unveiling with a formal launch the following year. When questioned about AI PCs in May, nVidia CEO Jensen Huang hinted that onlookers should ask the company again in 2025. MediaTek is reportedly collaborating on a 3nm SoC with the GPU giant, and Dell might also be involved.

Presumably, the processor would compete with Qualcomm's Snapdragon X Elite platform, which powers a lineup of Arm-based Windows laptops from various hardware manufacturers. Qualcomm developed Snapdragon X Elite with an exclusive license for Arm Windows processors, but Arm recently canceled licensing for all Snapdragon chips amid a rapidly escalating legal battle between the two companies. Furthermore, Arm previously said that more manufacturers, likely including nVidia, are set to enter the Arm PC market. Selling a consumer PC SoC represents a shift for Team Green, but it isn't entirely new territory. The company currently offers Arm-based data center chips and previously released a line of mobile processors, one of which powers the Nintendo Switch.

Aside from the range of vendors offering hardware, software remains one of the most important questions regarding Arm PCs. Apple has convinced the developers of many important applications to adapt to its shift to Arm with Apple Silicon to improve performance and energy efficiency, but it remains unclear if Windows can gain enough traction on the instruction set to do the same. Furthermore, consumer interest in AI PCs remains tepid. Customers aren't yet sold on AI applications, and some fear the emerging sector is heading for a crash resembling the dot-com bust.
Izvor: TechSpot

medo 01.11.2024. 22:34

Citiraj:

Autor Sinac (Post 3769962)
Raspberry Pi vrti Linux bez problema. Mislio sam općenito na ARM platformu koju netko želi iskoristiti kao zamjenu za klasično Windows računalo.

ARM PC/laptop bi trebao imati slične mogućnosti spajanja hardwarea i periferije te usporedive CPU performanse sa Intel/AMD platformom. RPi tu malo faila kao zamjena za desktop.

Bubba 02.11.2024. 14:50

Citiraj:

Autor medo (Post 3771667)
RPi tu malo faila kao zamjena za desktop.

CM-ove mozes stavljati u razne baseboardove ali opet tu ispadne skuplja pita nego tepsija za ne bas neki ekstaticni desktop uzitak. Cini mi se da cak niti nema CM-a baziranog na BCM2712 koji je u RbP5, tako da je ta kombinatorika dodatno neisplativija, jer te taj uzitak ispadne ~500$ za sto mozes dobiti respektabilni PC - daleko bolji od takve konfiguracije.

Sinac 20.12.2024. 12:22

https://www.nvidia.com/en-us/autonom...developer-kit/

https://www.bug.hr/racunala/nvidia-l...e-u-dlan-46009

Krenula nova runda ARM-a.

The Exiled 20.12.2024. 23:07

Citiraj:

Qualcomm defeats ARM’s claim over chip design license breach:kafa:
Citiraj:

Qualcomm Inc. prevailed at trial against Arm Holdings Plc’s claim that it breached a license for chip technology that the world’s largest maker of mobile-phone processors acquired when it bought a startup in 2021. Jurors in federal court in Delaware concluded Friday that Qualcomm didn’t violate the terms of an agreement covering Arm’s chip products acquired in a $1.4 billion purchase of Nuvia Inc. by incorporating the technology in its chips without paying a higher licensing rate. Jurors weren’t able to agree on whether Nuvia breached the license.

Qualcomm is one of Arm’s biggest customers and a longtime partner, but the companies have grown increasingly at odds as they’ve become rivals in the computer-processor industry. The dispute is important because many of the world’s biggest tech companies rely on chip architecture licensed from Arm and incorporated into Qualcomm’s products, ranging from computers to cars. Arm shares fell as much as 6.6% in extended trading following the verdict before recovering to trade down about 1.9% from their earlier close of $132.15. Shares Qualcomm gained about 2% in after-hours trading.

Arm claimed its Nuvia agreement should have been renegotiated after Qualcomm bought the startup and demanded the San Diego-based company destroy designs it got in the buyout. Qualcomm —- which is relying on Nuvia’s technology to push into the computer-processor market – argued to jurors it had a separate general license for Arm innovations that covered its work. UK-based Arm, which is majority-owned by Japan’s SoftBank Group Corp., sells chip designs and licenses a so-called instruction set — code used by software to communicate with processors. Qualcomm is the biggest chip provider to mobile phone makers such as Samsung Electronics Co.

The chief executive officers of both companies testified in the weeklong trial, which was filed in Delaware because Qualcomm is incorporated in the state, home to almost 70% of Fortune 500 companies. Delaware’s federal court also is a US hub of patent-infringement and licensing litigation.
Izvor: Bloomberg

The Exiled 29.12.2024. 17:45

Citiraj:

Samsung loses Qualcomm contract to TSMC for making next-gen Snapdragon chips:fiju:
Citiraj:

Samsung Foundry has once again failed to secure a contract from Qualcomm for the production of its latest flagship mobile processor, the Snapdragon 8 Elite Gen 2. This setback marks another chapter in Samsung's ongoing struggle to compete with industry leader TSMC, which has been awarded the entire manufacturing contract for the upcoming chip. The decision comes at a critical juncture for Samsung Foundry, which has been grappling with low yield rates in recent months. However, there are signs of improvement on the horizon. The company recently stabilized its 3nm production process, achieving acceptable yield rates, and is simultaneously pushing forward with the development of even more advanced 2nm chips. Samsung also has ambitious plans to produce 1.4nm chips by 2027.

Despite these advancements, Qualcomm's choice to partner exclusively with TSMC for the Snapdragon 8 Elite Gen 2 reflects the current state of the industry. TSMC's proven track record and reliability likely outweighed any potential cost savings that Samsung might have offered. This decision is particularly impactful given the widespread adoption of Snapdragon SoCs in flagship smartphones. The loss of this contract doesn't spell the end for Samsung Foundry's aspirations. The company is expected to make another bid for the production of the Snapdragon 8 Elite Gen 3 when it enters development. To improve its chances, Samsung will need to successfully fulfill orders from other clients and further improve yield rates.

This development coincides with reports of a significant price increase for the Snapdragon 8 Elite Gen 2, which could lead to higher costs for premium smartphones, including Samsung's Galaxy series. This situation presents a dilemma for Samsung's mobile division as it may need to consider alternative strategies to maintain competitive pricing. One potential solution could involve reviving Samsung's in-house Exynos chip line for future Galaxy devices, such as the S26 series. This move could help offset the impact of rising Snapdragon prices.

Alternatively, Samsung might explore partnerships with other chip manufacturers like MediaTek or make compromises in other components such as displays and cameras to offset increased processor costs. Despite these challenges, industry experts anticipate that Samsung's upcoming Galaxy S25 series will remain among the top smartphone choices in 2025. The company's strength in other areas of mobile technology should support its operations as it continues to refine its foundry capabilities and chip production processes.
Izvor: TechSpot

Sinac 29.12.2024. 19:25

Jel se to meni čini da je Qualcomm napravio dobar dizajn procesora na kojem mu čak i Samsung zavidi? :goood:

The Exiled 30.12.2024. 10:06

I to kaj veliš, makar je Exynos uvijek kaskal za Snapdragon ponudom, tak da ni ovaj razvoj situacije nije toliko čudan.
Citiraj:

Intel Foundry Services (IFS) and Arm today announced a multigeneration agreement to enable chip designers to build low-power compute system-on-chips (SoCs) on the Intel 18A process. The collaboration will focus on mobile SoC designs first, but allows for potential design expansion into automotive, Internet of Things (IoT), data center, aerospace and government applications. Arm® customers designing their next-generation mobile SoCs will benefit from leading-edge Intel 18A process technology, which delivers new breakthrough transistor technologies for improved power and performance, and from IFS’s robust manufacturing footprint that includes U.S.- and EU-based capacity.
Jedino ne znam kaj se desilo s ovom suradnjom koja je isto bila najavljena kao nešto kaj bude skratilo TSMC, ali izgleda da nije prošlo po planu i programu.:):D

The Exiled 14.01.2025. 16:42

Citiraj:

Qualcomm hires Intel's Xeon architect to lead development of server CPUs:kafa:
Citiraj:

Qualcomm has hired Sailesh Kottapalli, a former chief architect of Xeon server processors at Intel, to lead the development of the company's data center CPUs, reports CRN. Kottapalli spent 28 years at Intel and worked on various projects, including x86 and Itanium, as well as CPUs and GPUs. More recently, he was responsible for multiple generations of Xeon processors. Kottapalli joins Qualcomm as a senior vice president and will lead the company's data center CPU efforts.

Qualcomm's upcoming server CPUs are expected to use cores developed by engineers from Nuvia, a company Qualcomm acquired for $1.4 billion in 2021. Nuvia originally designed its Arm-based Phoenix (now called Oryon) cores with data centers in mind, but Qualcomm uses them for Snapdragon X system-on-chips (SoC) for consumer PCs, which upset Arm and led to litigation between the two companies. This is not Qualcomm's first foray into server CPUs. The company previously attempted to enter the market with Centriq but abandoned those efforts in 2018 and laid off its development team. This new initiative marks a return with a stronger focus and expanded resources.
Izvor: CRN

The Exiled 05.05.2025. 21:04

"Is x86 Actually Screwed?" ft. Wendell of Level1 Techs :sweating:

Sinac 06.05.2025. 11:58

Ne znam sjeća li se itko ovog iz 2017. godine, sad izgleda pomalo smiješno koliko se Intel batrga u bitci s ARM-om:

https://fudzilla.com/news/43862-inte...nst-x86-on-arm

Malo da proširim kontekst, Intel se našao ugrožen jer je postojao plan da se x86 instrukcije izvršavaju preko Win32 emulatora, čime bi se sve 32-bitne Windows aplikacije mogle vrtiti na ARM procesoru, nešto drugačije rješenje nego što je to današnji WOA operativni sustav. Intelu je zakipilo pod nogama jer je ova kombinacija ARM+Win32 app emulatora bila jednako brza ili čak i nešto brža nego na tadašnjim Intelovim mobilnim procesorima, barem ono što se pojavljivalo na raznim web stranicama.

The Exiled 06.05.2025. 12:23

Da, nažalost za Intel, Apple je svima dokazal da ARM itekako može proći, Qualcomm je uspjel i svoje posložiti, jedino još preostaje nadolazeća nVidia + MediaTek kombinacija. Intelove priče da ARM nije dovoljno dobar i općenito beznačajan su se kroz godine samo nastavile s rotacijom raznih direktora, pa su htjeli da se ARM čipovi proizvode u Intelovim fabrikama, mada od toga još nije ispalo nešto konkretno. Intelov (opet) novi direktor po uzoru na prethodnike odmah u startu najavljuje sve i svašta, a dobro bude, ako do kraja godine izbace jedan novi mobilni Panther Lake, a kamoli sva ostala čuda tehnike koja (na papiru) djeluju obećavajuće.

srale7 07.05.2025. 12:10

Citiraj:

Autor The Exiled (Post 3802848)
Da, nažalost za Intel, Apple je svima dokazal da ARM itekako može proći, Qualcomm je uspjel i svoje posložiti, jedino još preostaje nadolazeća nVidia + MediaTek kombinacija. Intelove priče da ARM nije dovoljno dobar i općenito beznačajan su se kroz godine samo nastavile s rotacijom raznih direktora, pa su htjeli da se ARM čipovi proizvode u Intelovim fabrikama, mada od toga još nije ispalo nešto konkretno. Intelov (opet) novi direktor po uzoru na prethodnike odmah u startu najavljuje sve i svašta, a dobro bude, ako do kraja godine izbace jedan novi mobilni Panther Lake, a kamoli sva ostala čuda tehnike koja (na papiru) djeluju obećavajuće.

Ovo je identična priča kao kada se onomad direktor u Microsoftu smijao iPhoneu i opcenito telefonima bez fizičke tipkovnice, jer je njima to u tom trenutku bilo nesmisleno.

Tehnoloska povijest nas je ako nista drugo naucila, da ako netko nije inovativan, da se mora na vrijeme orijentirati na novi smjer, jer u protivnom ce završiti kao NOkia i slični:D:D

Cini mi se da je Intel na dobrom putu da završi (ne mozda u tolikoj mjeri) slično

The Exiled 07.05.2025. 12:25

Intelovo bajno vodstvo voli ismijavati konkurenciju (1 - 2), posebice kad je već svima jasno da ih je ta konkurencija poprilično nadišla. ARM, AMD, nVidia, TSMC - svi redom su bili predmet Intelovog ismijavanja, a uredno su (pre)uzeli popriličan dio Intelovog sveukupnog tržišnog udjela, dok se Intel u međuvremenu vlastitom krivicom našel u nikad goroj situaciji u povijesti firme.

Sinac 07.05.2025. 12:47

Tak to ispada kad su drugi rođeni pod sretnom zvijezdom, a jadnom Intelu je crna mačka preša preko puta. Kad se samo sjetim Intelovog uvjeravanja kako je njihov 14++++ nm proces proizvodnje procesora vrh tehnološkog dostignuća i to ne znam koliko godina za redom. Čudo jedno kako su drugi igrači na tržištu uspjeli nešto postići i prestići Intel koji ih je sve do jednog omalovažavao.

srale7 07.05.2025. 13:02

To je praksa, da ne kazem ljudska priroda, kad si velik i jak, sprdas se sa malima. Uljuljas se, mali grizu, trzista/trendovi se mjenjaju i ti preko noci postanes "veliki igrac u velikim problemima.

The Exiled 07.05.2025. 13:04

Istu stvar sad rade i pričaju za 18A proces koji vrlo lako može ispasti kao i njihov 10nm izlet, pa su ga kasnije brže-bolje preimenovali u Intel 7, jer lakše je tak, nego konkretno imati nešto u rangu ili bolje od konkurencije.:) Tak da, obzirom na sve Intel aktualnosti, čisto sumnjam da bude se kod njih nekaj stvarno promijenilo, a primjera radi, Lisa Su je za svoje vrijeme u AMD-u, dosad ispratila sedam Intelovih direktora.:):D Da ne spominjem kak su stručLJaci u Intelu tijekom 80-ih imali priliku uložiti u TSMC, ali su ih potjerali i okarakterizirali kao totalno nebitnu firmu, da bi im danas TSMC proizvodil ono kaj Intel nije u stanju sam sprovesti u djelo.
Intel Arrow Lake ...
https://i.postimg.cc/tCP2qWX2/L4-Ie-PLirz-Z5860de.jpg
:frend:

Dottore 07.05.2025. 16:34

Citiraj:

Autor The Exiled (Post 3802848)
Da, nažalost za Intel, Apple je svima dokazal da ARM itekako može proći, Qualcomm je uspjel i svoje posložiti, jedino još preostaje nadolazeća nVidia + MediaTek kombinacija. Intelove priče da ARM nije dovoljno dobar i općenito beznačajan su se kroz godine samo nastavile s rotacijom raznih direktora, pa su htjeli da se ARM čipovi proizvode u Intelovim fabrikama, mada od toga još nije ispalo nešto konkretno. Intelov (opet) novi direktor po uzoru na prethodnike odmah u startu najavljuje sve i svašta, a dobro bude, ako do kraja godine izbace jedan novi mobilni Panther Lake, a kamoli sva ostala čuda tehnike koja (na papiru) djeluju obećavajuće.

Ako dobro sjećam, Qualcomm ARM prve generacije doživio flop glede preformansama i skupljim cijenama iako su kunili da je brži.
Mislio si na drugu generaciju ?

The Exiled 07.05.2025. 16:45

Mislil sam na ove sad aktualne Snapdragon X modele koji su u Windows ARM laptopima.:chears:
EDIT:
Citiraj:

nVidia N1X and N1 SoC for desktop and laptop PCs expected to debut at Computex:kafa:
Citiraj:

The N1X series is supposedly designed for desktops, while the N1 (without X) would target laptops. An official introduction is expected during either the MediaTek or nVidia keynote. The N1 series is said to be derived from the SoC developed for DGX Spark, a Mini-PC featuring an ARM-based CPU and a Blackwell-based GPU. The DGX Spark version includes a Grace CPU with 20 ARM cores: 10 Cortex-X925 and 10 Cortex-A725, and a Blackwell GPU capable of delivering 1 PFLOP of compute at FP4 precision. It also comes with up to 128GB of LPDDR5X memory, matching configurations seen in Mac Studio and certain Ryzen AI MAX systems, enabling support for large language models.

However, according to ComputerBase, the consumer version of the N1 may feature fewer cores, possibly 8 to 12, and reduced memory support, likely between 16GB and 32GB. As a result, N1-based PCs would be significantly cheaper, unlike the DGX Spark, which costs over $3000. nVidia is keeping specifications under wraps, and there are still more questions than answers regarding the N1 debut. Hopefully, nVidia has learned from Qualcomm’s Snapdragon Elite launch and avoids rolling out premium-only options backed by sponsored reviews, which only hurt initial reception.
Izvor: ComputerBase
Citiraj:

Qualcomm Snapdragon X Elite Benchmarks On Ubuntu Linux vs. AMD vs. Intel:fiju:
Citiraj:

Now approaching the one year point, the overall state of the Qualcomm Snapdragon X Elite support and performance is rather disappointing. Here's a look at where things currently are and performance relative to AMD Ryzen and Intel Core Ultra when making use of the latest Ubuntu Linux support. Treat this as just the preliminary benchmarks for the Qualcomm Snapdragon X Elite on Linux. As mentioned there isn't the CPU/SoC power monitoring ability at the moment under Linux so can't compare the X1E-78-100 to the AMD Ryzen and Intel Core power findings.

Overall the Snapdragon X Elite within the Acer Swift 14 AI came out just ahead of the Core i7 1185G7 "Tiger Lake" laptop and similar to the AMD Ryzen 7 PRO 5850U. Well behind the very newest AMD Strix Point (Zen 5) and Intel Lunar Lake generation parts. The Snapdragon X Elite on Ubuntu 24.10 and Ubuntu 25.04 did have some nice wins in the Firefox browser performance, some creator applications, and other workloads that are well tuned for AArch64, but overall was very mixed.

For most of the workloads, the AMD Ryzen AI 300 "Strix Point" and Intel Lunar Lake processors were delivering much better performance and they have also advanced nicely for showing x86_64 power efficiency. Yes, the Snapdragon X Elite is nearly a one year old platform now but still not even fully working under Linux and thus who knows how long until the Linux support is in good shape for the Snapdragon X Elite 2 / X2 Elite or whatever the next-generation Snapdragon laptop SoCs will be called. In any event it was an interesting initial experience with the Qualcomm Snapdragon X Elite on Ubuntu Linux and fairly straight-forward. Hopefully by the time of the Ubuntu 25.10 or Ubuntu 26.04 LTS releases the ARM64 desktop image will be in much better shape.
Izvor: Phoronix

Sinac 14.05.2025. 10:20

https://www.tomshardware.com/pc-comp...cs-at-computex

Baš me zanima što će biti prezentirano, pogotovo radi GPU-a koji bi trebao donijeti dobre performanse čak i u igrama.

The Exiled 14.05.2025. 11:31

IMHO hardver bude standardno sasvim solidan, a ukoliko se obesitne i ove AMD ARM glasine i najave - tim bolje, pa onda ostaje jedino Microsoft dio priče.:)
Citiraj:

AMD rumored to introduce ARM-based “Sound Wave” APUs by 2026; Expected to rival nVidia’s AI PC chip:sweating:
Citiraj:

AMD is reportedly developing an ARM-based CPU, the "Sound Wave" platform, which will be included in the Microsoft Surface lineup by 2026. As to what we can expect from the project "Sound Wave", it is reported that it will feature an RDNA 3.5 iGPU, along with a 3nm process from TSMC, similar to what NVIDIA is doing with its supposedly "N1X" SoCs. Microsoft has refined its "Windows on ARM" platform since it received massive adoption driven by Qualcomm chips, so with AMD entering the segment, we can see it growing even further. Interestingly, Sound Wave won't be AMD's first venture with the ARM platform, as back in 2014, the firm introduced Project Skybridge, which was said to combine x86 and ARM in a common platform, but was later cancelled out due to economic and market adoption concerns. We are not certain when AMD will showcase its ARM-based APUs for now. However, the SoCs will be featured in Microsoft's Surface laptops by 2026, so the official unveiling could likely occur either at the end of this year or maybe at CES 2026.
Izvor: Interneti

tomek@vz 18.05.2025. 17:29

Citiraj:

It's been nearly 20 years since Apple launched the Microsoft-mocking "I'm a Mac, I'm a PC" commercials, and the two companies continue to take shots at each other in their respective ads. The latest from the Redmond giant claims that top Copilot+ PCs are more than 50% faster than a Mac – specifically, the now-discontinued MacBook Air M3 that launched in March 2024.
Microsoft's 12-second ad, which doesn't exactly scream high-end production values, was shown on the official Windows YouTube channel. The announcer excitedly boasts that "Top Copilot+ PCs are up to 58% faster than MacBook Air with M3." Just to ensure the point is driven home, he adds, "That's right, faster than a Mac."

> TechSpot


Zbilja ne kužim kak im se da lagat.

Sinac 25.05.2025. 12:02

Situacija na tržištu postaje zanimljiva:

Xiaomi XRING O1

https://www.mi.com/global/discover/article?id=4926

https://www.theverge.com/news/672541...ro-pad-7-ultra


Ovo bih najradije i u Intel thread postavio, nadam se da direktori Intela čitaju ovakve vijesti, no, ionako ih je od prvog do zadnjeg vrijeme pregazilo...

The Exiled 25.05.2025. 15:51

Za Intelove direktore je uvijek najbitniji onaj sljedeći kvartal i tak već posljednjih deset godina, od kvartala do kvartala, od direktora do direktora, od rekordnih zarada do rekordnih gubitaka. Nema tu pomoći, oni se i dalje prave da će se sve nekako samo od sebe riješiti, a najbolje bi bilo da sve firme koje nisu Intel, jednostavno prestanu s čim se već bave, pa postepeno nestanu i ostave Intel na miru, jer ništa Intelu nije falilo dok se ostali nisu počeli miješati u Intelov dobro uigrani sustav.:):D

Sinac 25.05.2025. 18:03

Je, istina, baš si dobro zaključio da je priča išla od kvartala do kvartala. :lol2:
A najbolja stvar je što će ovi energijski niskozahtjevni procesori tek doći do izražaja i Intel će suze roniti što svojim Atom procesorima ne može parirati ostatku tržišta.

The Exiled 25.05.2025. 18:10

E, nažalost Atom nisu stigli ili mogli dobro unovčiti čak i kad su na početku te priče imali neku šansu. Kasnije su Atom dio brže-bolje pretvorili u tzv. šparne jezgre, odnosno E-core implementaciju, jer je svojevremeno to bilo jedino rješenje da konkuriraju šrot AMD-u. I makar sasvim solidno, opet nije bilo dovoljno dobro, bilo za ARM, bilo za AMD konkurenciju.

Sinac 21.07.2025. 10:07

Mislim da će i AMD i Intel morat držat gaće kad se Nividijin ARM laptop pojavi na tržištu. Za sada su rezultati bencheva dosta dobri, još ovisi kako će se grafička kartica ponašati:

https://www.club386.com/nvidia-arm-c...zen-7-9800x3d/


Sva vremena su GMT +2. Sada je 23:18.

Powered by vBulletin®
Copyright ©2000 - 2025, Jelsoft Enterprises Ltd.
© 1999-2024 PC Ekspert - Sva prava pridržana ISSN 1334-2940
Ad Management by RedTyger