AMD did they Mess it up for us?

[Kumagoro]

In the long run is it good that AMD made a 32/64 bit chip? Personally I wish that intel would have been able to design a new modern 64 bit chip which didnt have to have support for old x86 instructions. Thus making a faster and more effcient PC.

Granted you might not be able to have backwards compatability even with emulators but I would rather have something which is the best it can be and not held back.

Will we ever get a new modern chip or is it too late? it feels like an excellent opprtunity was missed.

[turkster]

Yeah but migrating to hardware that was exclusively 64bit would be fraught with difficulties. It would requirte a completly new OS from day one otherwise it would be useless. Also the software progs would asll have to be redesgned to be compatable. Undoubtedly this would not happen overnight so the take up of the new hardware would be very slow.
With 32/64bit systems migration will be a lot smoother and proabably quicker.

[nichomach]

In the long run is it good that AMD made a 32/64 bit chip? Personally I wish that intel would have been able to design a new modern 64 bit chip which didnt have to have support for old x86 instructions. Thus making a faster and more effcient PC.

Granted you might not be able to have backwards compatability even with emulators but I would rather have something which is the best it can be and not held back.

Will we ever get a new modern chip or is it too late? it feels like an excellent opprtunity was missed.

They did; it's called Itanium(which only supports x86 through emulation), and with the best will in the world, even in its supposed target market of heavy application servers and scientific applications it's hardly flying off the shelves. And it's not as if it wasn't supported by an OS from day one, at least on the Unix side.

Actually, I think what's happening is quite healthy; 64-bit will become the de facto standard, but it will happen gradually, and the mass of applications out there will catch up to use it gradually.

[Stringent]

Remember when we went from 16-32 bit. The process was the same old slow process of moving things over to it.

[Lee H]

AMD did us a good thing as Intel we're dragging their feet releasing the next-gen of processors so for AMD to get the glory and release a 64 bit processor before Intel was a real kick in the trouser spuds for the top dogs at intel.

have you noticed now all the roadmaps at intel are now chopping and changing, with 64 bit support gaining more priority and also the fact they are now giving their processor "performance" numbers - 530, 532 etc is just a laugh as far as I am concerned as they slagged off AMD for this roughly 2 years ago didn't they

Its all swings and roundabouts in this industry guys

Next step for intel is probably going to be a multi-cored CPU but I hope AMD beat them to this again and show them that their is no place for complatency within the IT industry on a whole. Take a look @ Nvidia - they are now launching SLI again !!! ( all the 3dfx fanboys are weaping now praising the silicon god above for this day )

Whatever the future holds, its going to be a fun 12-24 months ahead

[Stoo]

nichomach hit the nail on the head..

The Itanium died on it's feet, too expensive and far too slow, Intel cranked the performance up with the Itanium 2, but it's still slower than a low end Opteron system, but far, far more expensive.. and that's running in a native 64 bit linux environment.

[Stoo]

IIRC AMD are taping out their first dual core chips

[Kumagoro]

I think it was inevitable that there would be a 32/64 bit transition. Maybe when everything has settled down a new viable platform with be available....... hmm nope it will never happen.

[Lee H]

IIRC AMD are taping out their first dual core chips

looks like a case of the David beating Goliath yet again

I have nothing but respect for AMD, first they moved to the hypertransport system which is a great piece of design, and hopefully the new dual cored processors will give us gamers the performance we desire

[Stoo]

Apparently "Nocona" is full of compatibility problems too..

This is an error message from a 64bit RH distribution when it detects a Nocona chip:

“Software IOTLB — Intel® EM64T does not support an IOMMU in hardware while AMD64 processors do. This means that physical addresses above 4GB (32 bits) cannot reliably be the source or destination of DMA operations. Therefore, the Red Hat Enterprise Linux 3 Update 2 kernel "bounces" all DMA operations to or from physical addresses above 4GB to buffers that the kernel pre-allocated below 4GB at boot time. This is likely to result in lower performance for IO-intensive workloads for Intel® EM64T as compared to AMD64 processors.”

oops

[Butuz]

AMD conentrated in making faster processors running on better architecture. Intel conentrated on slagging amd off, making better TV adverts and then copying AMD (sounds kinda like Labour, and Conservatives does it not?). That is the reason we are where we are today!

Butuz

[DR]

AMD did a cracking job - their CPUs brought real performance to the masses, 64bit computing doesn't matter at the moment and needed to be launched to pave the way.

Intels new Xeons (nacona) are its first "x86-64" Extended Memory 64 Technology CPUs to ship. The top-of-the-line 90nm Xeon is clocked at 3.6GHz with an 800MHz FSB, is hyperthreaded, and contains 13 new Streaming SSE instructions and Speedstep. Most notably, the 90nm Xeon runs both 32-bit and 64-bit software. The first batch of new Xeons is targeted at dual-CPU workstations with server versions due out in the next two months. Processors designed for use in four-way (or more) configurations will be available late this year or in early 2005.

So how does the new Xeon stack up against the AMD Opteron? The one advantage it has is its ability to utilize DDR2, which Opteron cannot. On the other hand, the Opteron has an integrated memory controller and consumes less power than the Xeon (we will see if that remains the case once the 90nm Opterons are shipping later this year). Along with the new Xeon, Intel is also launching the E7525 chipset (Tumwater), which features support for PCI Express and the aforementioned DDR2. iWill and Tyan have both announced new motherboards for the Xeons based on the E7525.

AMD has gained some momentum in the server space with the Opteron, with several OEMs, including Sun and HP, offering workstations and servers with AMD's flagship CPU. The success of AMD's x86-64 architecture forced Intel's hand, leading to January's news that Intel would be offering their own 64-bit Xeons. Intel's last set of 32-bit Xeons released in March closed the gap on the Opteron, mostly due to the hefty 4MB of L3 cache (benchmarks for the new Xeons are not yet available). In addition, the new Xeons are priced fairly reasonably, at US$851 in quantities of 1,000. What remains to be seen now is how well the Itanium will fare now that the 64-bit Xeon is available. While Intel is confident that the new 64-bit Xeons will not cannibalize Itanium sales, sales of the Opteron have demonstrated the market's desire for a relatively-inexpensive 64-bit CPU. The new Xeon is aimed squarely at carving out a big slice of that market for Intel.

[Dorza]

AMD have done good without a doubt. They are driving technology forward. Intel being the market leader wouldnt have been so quick to get 64-bit to the desktop like AMD have been. Now all of a sudden their going to release a prescott with is 32/64bit compatiable some time soon iirc. To further show how much of a kick in the teeth Intel have sustained and their want to rush out 64-bit they are actualy copying AMDs CPUS and useing the design in their chips (so ive read). Its no longer the other way round. To me this is the sign of a desperate company.

[Rythmic]

It is a pity that we seem to be sticking with the same ancient and rather nasty CPU architecture thats been patched time, and time again. Apple managed to get away from 68000 to PowerPC.

That said, I've never really liked Itanium either.

It's x86 emulation is incredibly slow (a good p3 can beat them) which left buyers without an easy upgrade path for existing software, and it requires an extremely complicated compiler to get any performance out of them at all. With the early versions (pre madison) being slow and the compilers not up to spec, it didn't make a good early impression.

AMD 64 looks good for enterprise buyers though - they don't have to recompile their code, but if they do, it's a fairly simple procedure that has no architecture changes to look out for, and allows them to take advantage of double the registers and most importantly >2GB per thread without any of that nasty AWE business.

64 bit address space is whats important here - not so much 64bit processing.

It's a speed boost (and RAM increase) without too much bother - which is what the buyer generally wants in these days of much reduced IT budgets.

It's no wonder that Intel is being forced to come out with a competitor. It's just a pity we don't seem to be able to shed the x86 CISC instruction set (yes I know the processors are themselves no longer CISC - but the instruction set still is).

Ironically, for a bit of brit pride, Intel sell another architecture - one which they have to license, but it sells many more units than any of theirs.

That'll be ARM

[Kumagoro]

I guess what AMD did had to be done because thats the way the PC market works. When 64 bit computing is up and running properly i.e. when windows 64 comes out, will a new platform be possible or will windows64 effectively stop that happening, because of the way its designed?

Hasnt intels position been forced by Microsoft because they said they wouldnt do different versions of a 64bit windows?

Will we forever have the saggy arsed legacy that is x86 cisc? it seems analogous to the way windows was stuck with DOS etc.

another question anyone heard anything recently about the new bios system intel was supposed to be designing? I havent seen anything on it for over a year.