ARM does have a huge cash pile post IPO. So they can afford to keep this going until all the avenues are exhausted.
The problem is that this was their best shot at it and they fumbled the bag completely.ARM does have a huge cash pile post IPO. So they can afford to keep this going until all the avenues are exhausted.
Yes, emulation; but I wanted to emphasize the accelerating cadence of support and adoption for ARM in that ecosystem by the volume of improvements that recently made their way into software and hardware.You misspelled emulation 😑
It's better than nothing, but:
#1. It's imperfect compatibility at best.
#2. Emulation comes at a cost to power and/or performance vs native code.
When battery constrained mobile markets are the main focus of WoA this is not a trivial issue.
This is precisely why Apple set a firm date to phase out Rosetta2 support, to force devs to get their posteriors in motion sooner rather than later.
Infortunately this isn't something Microsoft can really do - best they or others can do is throw money at the problem and hire extra programmers for each dev to speed up their efforts, aka the nVidia approach.
Wasn’t the 8 Elite (Oryon v2) the first clean sheet? Certainly was a significant improvement from the X Elite (v1).Yes, emulation; but I wanted to emphasize the accelerating cadence of support and adoption for ARM in that ecosystem by the volume of improvements that recently made their way into software and hardware.
There is a also a new, pervasively used ARM64EC ABI as well which better enables drop in support when code to be ported has dependencies on legacy, unmaintained x86 binary blob plugins. Due to the MO of MSFT, this support is absolutely critical vs say iOS where fat binaries sufficed due to the regular transitions enforced by fiat: this is a complexity inherent to this MO which MSFT is handling about as well as anyone could. I agree it is an albatross around the neck of performance, but this legacy support is a key feature of Windows and performance is good enough for most clients who do need this support and there is now an easier path to transitioning to fully native binaries.
SX Elite definitely helps fill in part of the emulated performance gap, and while not at the level of native code on Macs yet, it is trading blows with x86 parts even in emulation and that's all it needs to tear away a chunk of the Windows platform for now. Future high performance / high efficiency SoCs from multiple vendors will further help fill in that gap, and next year's Qualcomm big cores will be the first clean sheet design by the former Nuvia team aimed at the client space.
My understanding is that it's Oryon with tweaks like DVFS for mobile and a cache shrink without much penalty; small cores were also introduced. Apparently v3 is code named Pegasus and supports ARM v9.Wasn’t the 8 Elite (Oryon v2) the first clean sheet? Certainly was a significant improvement from the X Elite (v1).
It's also a physdes pass since the og Phoenix implementation in Hamoa looked very rough.My understanding is that it's Oryon with tweaks like DVFS for mobile and a cache shrink without much penalty
From what you have said they have flooded the retail channels, which is not the same thing as actual sales and long term retention due to software compatibility or performance issues.Those new QC ARM cores will be killer.
QC CEO said they have already captured 10% of the market. Am I surprised? Nope, I actually believe him. The QC laptops are everywhere in Australia. JBhifi, Harvey Norman, Officeworks, good guys. Practically the Australian equivalents to best buy.
They flooded the markets worldwide, not bad
Trying to sue your biggest customer does not seem like a good idea in the first place.Good, stupid ARM. What a fail, lol.
Well, maybe there would be no other choice but not in this case, no.Trying to sue your biggest customer does not seem like a good idea in the first place.
Naw...Sorry to resurrect an old thread, but:
It's over, Qualcomm is off the hook.
They've been ridiculed already. What's more to lose ?Naw...
"(...) a jury finding that Qualcomm's personal computer chips are properly licensed under its license agreement with Arm.
The jury, however, could not reach a unanimous verdict on other questions presented at the trial, and Arm has filed a motion for a new trial."
Arm is intent to keep entertaining its and Qualcomm's lawyers some more.
Right but that pertains to other issues raised in the suit. The core issue of the transferability of the Nuvia license is no longer disputed.Arm is intent to keep entertaining its and Qualcomm's lawyers some more.
I'm not so sure it's as clear cut as you all state.
Arm is all about extracting money of its licensees nowadays, the biggest difference between v8 and v9 was the essentially doubling the license fee licensors have to pay.
As for the court case my understanding is that Arm tried to push for its biggest possible leverage it has to extract money, that is breach of contract and as such nullifying existing agreement that earned Arm less (= "too little") fees. Both Nuvia's and Qualcomm's individual licenses were targeted and older and as such something Arm was intend to replace with contracts earning them higher fees.
While the breach of contract accusation is now resolved, you will still see Arm trying all possible other avenues to still extract the value from Qualcomm they feel entitled to.
Arm has shown to not care expect for money, these court cases are the result. It would be much better for Arm licensees if Risc-V were already much more feasible and competitive yesterday. Softbank and the whole prolonged failed Nvidia takeover turned Arm into a different company.Clearly. But spooking your customers isn't really that smart of an idea IMO.
RiscV clearly isn't a competitor, yet.
But being RiscV or another, the moment there's another possibility, Arm will have their bottom getting hot very quickly.
Xiangshan is that since it is paired with CRVA, China RISC-V Alliance (中国开放指令生态联盟). With Tsinghua Univ., Peking Univ., Baidu, Alibaba, SMIC and Unisoc (Unigroup Spreadtrum & RDA), Huawei, Tencent. China RISC-V Industry Consortium, CRVIC is the other group: VeriSilicon, Espressif Systems, Ingenic Semiconductor, UNISOC (Spreadtrum & RDA), Zhaoxin Semiconductor, Horizon Robotics, and C*Core Technology. This(CRVIC) is the Alibaba, Nuclei, StarFive one that has: "agreed to form the alliance that includes a condition for members to not sue each other over patent infringement"RiscV needs the consortium treatment. It needs an alliance of CPU producers/manufacturers etc. to all come together, sign the absolute tightest IP sharing and hold harmless agreements known to man for the product of their efforts, and get some sort of functional consistency and completeness for the ISA. I realize that there are published levels to RiscV already, but, it's still waaaaaay too flexible of a target.
It's likely the suit against Nuvia and Qualcomm was intentionally frivolous and an attempt to stall what ARM considers a likely competitor in its new business of implementing ARM designs as well:Arm has shown to not care expect for money, these court cases are the result. It would be much better for Arm licensees if Risc-V were already much more feasible and competitive yesterday. Softbank and the whole prolonged failed Nvidia takeover turned Arm into a different company.