Frickwhen Core 2 had made its splash felt, I can imagine you getting an x2 3600+ for something like that if it was on sale.

It was before Conroe, when AMD reigned supreme.

medi01Read it again and think what 2004-5 could mean.

I know what 2004-5 means. If that is indeed true, you would have had to have bought them right when they were released at mid to end of 2005, and you definitely didn't get the motherboard RAM and processor for $220.

medi01Read it again and think what 2004-5 could mean.

I repeat, for 220 Euros, I've got dual core Athlon 64, new mainboard and mem. (don't remember amount though :S)

I don't remember what date it was, Wow released at least 1 year before that, roughly at that time. Wow was the reason I was upgrading.

Mobo was about 55Euro, CPU under 100, maybe as low as 80.

My gmail can't look back that far, or else I'd had exact numbers.

First of all, you wrote "220$". $ =/= €
$ means Dollars
€ means Euros

Now you're just changing your story to try to save face. But even if we go with the idea that you meant Euros, €220 wouldn't buy you a Athlon X2 a motherboard and RAM. At least not at retail prices. Maybe if they fell off the back of a truck. The exchange rate for the Euro back in 2005 wasn't that strong. It was about €1.25=$1. So even if you bought the cheapest 3800+, that would still be €280 just for the processor. So there is no way, at anytime in 2004-5 that you bought a Athlon X2, motherboard, and RAM for €220.

The fact is, as soon as AMD was on top, they started charging crazy prices for their processors. Up to $1,000 for non-Extreme processors even! And it didn't go back down again until Intel released the Core 2 series and topped AMD again.

medi01I recall it was before first WoW expansion came, so, Burning Cruzade, released in Jan 2007.

That would have been late 2006-early 2007. So, go back and read again, and think really hard what 2004-5 means.

And at that point, Intel had released their Core 2 line. Which, as I said, topped AMD and caused them to drastically lower the prices.

medi01It was before Conroe, when AMD reigned supreme.

Ok, seriously, you gotta get your story straight. If it was before Conroe, then it was WAAAAAY before Burning Crusade. And if it was before Conroe, then the processor were super expensive. Unless you got it at that very short time right before Conroe came out, when all the venders were slashing X2 prices just to clear stock before Conroe's launch. But even still, that doesn't prove the original point you were arguing, that AMD didn't jack up their prices when they were in the lead.

xfiawhat is this bullock of half information.. do you work for nvidia? haha

I linked directly to TPU's own review.

LemmingOverlordI won't lie to you, I haven't benchmarked an FX-8350, much less an overclocked one, but I've benchmarked other CPUs of the same architecture and extrapolating results (to me), and checking on benchmark sites, there is no chance the 8350 comes within that 10% window - clock for clock in multithreading - unless you are quoting one very specific benchmark (that's the problem with benchmarks, right?). It does come down to what benchmark you are running and whether the system is somehow bottlenecked by something... i.e.: do you have to flush something to disk? (from your sig, you have overclocked your FX-8350 and it's got 32GB of RAM which might play a role in this). Clock for clock, single or multithreading, Haswell beats Piledriver...

ALSO: upping IPC by 40% does definitely not mean they are increasing single-threaded performance by the same ratio. In order to get IPC up 40% they will surely rein in the CPU clock. AMD is doing to its processors what Intel did when it about-faced on Netburst: it's improving execution (reducing pipeline length, branch prediction, etc... we'll see when it surfaces).

ALSO: Intel is not frozen in time, right...? It'll be a little over 2.5 years between the launch of Zen and Haswell (if Zen is on time)... Concern yourself with the fact that AMD has not been making any progress in CPUs for the past couple of years, while Intel integrated graphics have been steadily improving, generation after generation.

Don't get me wrong: I want Zen to succeed, but AMD has so far given me nothing but reasons to doubt their promises/expectations. They have systematically fallen short on delivering the gains they promise, time and again.

If things go pear-shaped, CES 2017 will be a very - believe me - VERY public place to crash and burn. Hence my "doom-and-gloom prediction".

There should be a betting pool for this... Anyone?

Here is a link to a site with full sets of benchmarks on both the i7 4790K and the FX-8350

heres some benches from my 8350 at stock and at 4.5ghz with a comparison on the i7-6700k
13% difference multi thread at stock and the 8350 is afew Years older, also my system has a lot of dev stuff running on it plus mssql many instances etc

Stock AMD 8350 4Ghz vs i7-6700k 4ghz stock


AMD 8350 with quick oc to 4.5ghz



if zen really has 40% ipc over pd it will monster intels current lineup, single thread is a weak point.

thought I better add the single thread is approx. 40-42% behind the i7-6700k in this quick bench, so if the ipc was on single thread it would be on par with current gen intel and multi thread if improved and I cant imagine why it wouldnt be beyond intels current gen.

medi01It was before Conroe, when AMD reigned supreme.

Then it couldn't have been that low, unless you bought it for way less than the retailers did.

DigitalUKheres some benches from my 8350 at stock and at 4.5ghz with a comparison on the i7-6700k
13% difference multi thread at stock and the 8350 is afew Years older, also my system has a lot of dev stuff running on it plus mssql many instances etc

Hey. Well, like I said, it depends on what benchmarks you're using, the link at the bottom of my other post shows 4790K benchmarks on a number of uses (both gaming and non-gaming) trouncing anything from AMD. I am not too familiar with the CPU-Z benchmark, but according to it my 3770k just obliterates a 4790k, while being clocked 500MHz lower. Still, it's a synthetic benchmark and most end-user apps, despite multi-threaded, are still far from optimized...

Anywho. I eagerly await the release of the Zen dreadnought upon the unsuspecting masses. It will be a make-or-break deal for AMD, and that is what ticks me off, no-one in the industry is looking at Zen with the notion that it will either catapult AMD into the mainstream again, or kill it outright..

The more a "build-up" a brand has to a product, the worse it will turn out... either because of our own exaggerated expectations, or because the marketing BS will just cherry-pick scenarios and applications to underline a very specific gain in performance and give the public a sense of "overall" performance gains. If Su starts talking more and more at investors on the gains, and benchmarks start "leaking"... I'll be concerned.

FrickThen it couldn't have been that low, unless you bought it for way less than the retailers did.

Yes it was, 220 total price for mobo + cpu + mem.
At the same time my guildmate built "fastest (non EE) Intel you can get" PC and it was a Prescott.

medi01Yes it was, 220 total price for mobo + cpu + mem.
At the same time my guildmate built "fastest (non EE) Intel you can get" PC and it was a Prescott.

Yeah, I think we've established there was no way that is possible.

Intel Released the E6600 Jul-2006 for $316. Proof
Intel Released the E6700 Jul-3006 for $530. Proof
AMD Released the AM2 FX-62 May-2006 for $1031. Proof
AMD Released the AM2 X2 5000+ May-2006 for $696. Proof
AMD Released the AM2 X2 4400+ May-2006 for $514. Proof
AMD Released the AM2 X2 3800+ May-2006 for $323. Proof

This shows what AMD's prices were just 2 months before Conroe was released. So there is no way you bought an X2, motherboard, and RAM for 220 € or $ before Conroe's launch.

The other interesting thing I'd like to point out is that everyone rags on Intel for keeping prices high. But even though the E6600 was faster than 5000+, and basically just as fast or slightly faster than the FX-62, they priced it at $316. That is cheaper than even the X2 3800+, and it crushes that processor. It is actually Intel that forced AMD to lower their prices from ridiculous levels to more something more reasonable.

Yeah, it was Sept 2007, 9 month into BC, mia culpa.

Here is the list:

Bezeichnung : Athlon64 X2 4000+
Einzelpreis : 63,55
Gesamtpreis : 63,55
Anzahl : 1
Bezeichnung : Ultra DIMM 2 GB DDR2-800 Kit
Einzelpreis : 74,34
Gesamtpreis : 74,34
Anzahl : 1
Bezeichnung : M2A-VM
Einzelpreis : 50,68
Gesamtpreis : 50,68

It actualy was 188.57 Euro total for GPU + Mainboard + Mem.
I bought it from MIx Computers de, if that matters.

newtekie1AMD Released the AM2 FX-62 May-2006 for $1031.

newtekie1even though the E6600 was faster than 5000+, and basically just as fast or slightly faster than the FX-62

You conveniently forgot Pentium Extreme Editions, that have actually established the 1k$ price point, that FX-62 thoroughly wiped the floor with.



www.anandtech.com/show/2012/9

medi01Yeah, it was Sept 2007, 9 month into BC, mia culpa.

Here is the list:

...

So now it has gone from 2004-5, to sometime in 2006 but definitely before Burning Crusade, to it was definitely before Conroe, to now late 2007. So again I ask you, what you asked me, do you know what 2004-5 means?

So, yeah, completely bullshit, like I said. And by the way, Sep. 2007 was a long while after Conroe was released. I went over this, once Conroe was released AMD had no choice to slash prices. The E6600 beating their $1000 5000+ and only being priced at $316 forced them to drastically cut prices. In fact, you bought your Athlon X2 after Allendale was released, with the E4300 besting the X2 4600+, and biting at the heals of the 5000+. All for $163 at launch. So no wonder you got an X2 4200+ so cheap.

You're cost to buy an Athlon X2 when AMD was not in the lead doesn't help your argument that AMD didn't jack up prices when they were in the lead.

medi01You conveniently forgot Pentium Extreme Editions, that have actually established the 1k$ price point, that FX-62 thoroughly wiped the floor with.



www.anandtech.com/show/2012/9

No, I didn't forget them. I clearly said the EE is a niche market that is always overpriced. Intel might have set the $1000 price point, but AMD is the one that priced non-Extreme processor at that $1k price point. However, Intel's prices in the mainstream market are, and have been, reasonable.

newtekie1So now it has gone from 2004-5, to sometime in 2006 but definitely before Burning Crusade, to it was definitely before Conroe, to now late 2007. So again I ask you, what you asked me, do you know what 2004-5 means?

To both of you, not singling out anyone: I just want to point out this has gone a bit off-topic (we're no longer debating the specs on Zen and its potential).

Let's move on? mmmmkay? :)

newtekie1Intel might have set the $1000 price point, but AMD is the one that priced non-Extreme processor at that $1k price point

So, AMD is guilty of pricing CPU that was FASTER than competitors 1k$ CPU whopping 1031$. Mind boggling.

4000+ that I bought in Sept 2007 for 74 Euro was released in May 2006.

LemmingOverlordLet's move on?

Fair enough.
Just checked the "latest FX" vs i5 review. So, what we have now:

1) Intel has single thread IPC lead of about 60% (i7)
2) Intel 14nm Perf/Watt is about 2.2 times better than AMD on 28nm

If AMD achieves 40% IPC increase and Samsung's 14nm allow to cut power consumption in half, we can see competitive low and mid range AMD CPUs. However, AMD still hinting at "even moar coars" makes me worry, they actually won't... =(


PS

newtekie1By that logic, Intel should have...

Jeez Christ.... Here is the 999$ Conroe:
en.wikipedia.org/wiki/Intel_Core#Core_2_Extreme

medi01o, AMD is guilty of pricing CPU that was FASTER than competitors 1k$ CPU whopping 1031$. Mind boggling.

By that logic, Intel should have priced the E6600 or E6700 at over $1,000 too, and the E4300 at over $300. But they didn't. They easily could have, and kept prices high, but they didn't.

medi014000+ that I bought in Sept 2007 for 74 Euro was released in May 2006.

I don't see why that matters.

LemmingOverlordLet's move on? mmmmkay? :)

Sounds good.

DigitalUKif zen really has 40% ipc over pd it will monster intels current lineup, single thread is a weak point.

thought I better add the single thread is approx. 40-42% behind the i7-6700k in this quick bench, so if the ipc was on single thread it would be on par with current gen intel and multi thread if improved and I cant imagine why it wouldnt be beyond intels current gen.

zen is ment to be 40% ahead of their latest apu, not the older 8350. based on those loose figures zen will beat skylake clock per clock.

Parn40% IPC improvement over A10-7870K. Hope this is true as it will definitely bring Zen on par with Haswell at minimum.

40% IPC improvement over Excavator. Based on the actual design of the CPU. Everything else such as the Process Node is extra. I would assume we are looking for as much as 60% if not more versus Excavator.

ZEN should easily compete with Intel's future Gen CPU's, not just Haswell. Skylake no problem and even Cannonlake, which is simply a die shrink of Skylake.

vega22zen is ment to be 40% ahead of their latest apu, not the older 8350. based on those loose figures zen will beat skylake clock per clock.

This 40% IPC improvement over Excavator is independent of the process node. That alone is a testament, Skylake may have a difficult time keeping up with ZEN.

Super XPThis 40% IPC improvement over Excavator is independent of the process node. That alone is a testament, Skylake may have a difficult time keeping up with ZEN

Let's hope, but news about 32 core chips likely mean that single thread performance is not that great.

medi01Let's hope, but news about 32 core chips likely mean that single thread performance is not that great.

They'd have to reduce the clocks so ... obviously.

medi01Let's hope, but news about 32 core chips likely mean that single thread performance is not that great.

For clarity's sake: there are no "32 core" chips, there is a dual 16-core design with a new interconnect which brings everything together. This will probably look like the quad-cores of yore (2x2 instead of 4x1).

Also, 16-core CPUs will certainly have lower clocks than your run-of-the-mill desktop Zen, and the way to make up for that in terms of performance (without changing the chip architecture) is to introduce copious amounts of L3 cache, which is exactly what Intel does with Xeons.

From what I've read around this wonderful web:

• the Zen APU parts lack any L3 cache (meaning they'll be roadkill for Intel on single core performance). While I understand the strategy, it means that the APUs will fight an uphill battle, focusing on price rather than on performance.
• the Zen server parts are expected to carry 8MB of L3 cache per 4-core clusters, meaning a 16-core processor would sport 32MB L3 cache. A little under Intel's Xeons, but still a worthwhile performer if the performance AMD claims isn't BS.

Another of my concerns about Zen is "what niche is it addressing"? By the time Zen is out, Intel will have released its next-gen Grantley-EP Xeons (ES-46xx v3) which will up the ante to 22 cores (44 threads).

It's never to late in this industry. Once ZEN is release this yes it will eventually replace AMD'S entire line of Desktop CPU's and APU's.

And why are people worried about Single Threading Performance on ZEN? ZEN is not Bulldozer. ZEN should have the ability to outperform it's competition in both Multithreading and Single Threading real world performance.

ZEN I am brand new design built by one of the world's best CPU Architect.

Super XPAnd why are people worried about Single Threading Performance on ZEN? ZEN is not Bulldozer. ZEN should have the ability to outperform it's competition in both Multithreading and Single Threading real world performance.

People will stop worrying about single thread performance on Zen, the day AMD actually delivers what it promises.

IMHO: When a company overstates its CPU features and performance like it has repeatedly done, for the sole sake of a misguided product strategy which is framed by financial KPIs and exec bonus payouts, any expectations on Zen's final performance is strictly BS.

idk what you guys are going on here but zen wont be faster per core but it will probably be better at multithreading and the lack of l3 means nothing at all with the arch

xfiaidk what you guys are going on here but zen wont be faster per core but it will probably be better at multithreading and the lack of l3 means nothing at all with the arch

Only APU will lack L3 because the space is needed for the GPU, afait(heorize).

LemmingOverlordPeople will stop worrying about single thread performance on Zen, the day AMD actually delivers what it promises.

IMHO: When a company overstates its CPU features and performance like it has repeatedly done, for the sole sake of a misguided product strategy which is framed by financial KPIs and exec bonus payouts, any expectations on Zen's final performance is strictly BS.

Bulldozer was BS. That Design was the one they over stated.
ZEN is a different Story. AMD is not foolish enough to pull off the same thing. ZEN is life and death of this company. That's why they are not BS'ing.

Caring1AMD still playing catch up.
A 40% improvement in IPC will still leave them behind Intel.

Not necessarily. That 40% improvement is independent of the process node. It's based on the physical design of the chip clock for clock. The process node and any other chip/ chip-set improvements will be added to that 40% gain.
Theoretically, we could be looking at 50% to 60% IPC improvement, if not more.

medi01Let's hope, but news about 32 core chips likely mean that single thread performance is not that great.

There's no basis to your reasoning it seems. 32 Core Chips or Multi-CPU platforms are needed for Servers and Workstations. I don't believe we will see 32 Core Chips for performance desktop setups.
What interests me is ZEN's Quad-Channel memory interface. It's about bloody time. Based on what I've read about ZEN, this will surely be my next major upgrade.

Super XPTheoretically, we could be looking at 50% to 60% IPC improvement, if not more.

Can't really say that because IPC is short for instructions per clock, so you can't count in potential clock difference and still call the metric IPC.
Nevertheless, I predict Zen won't be clocked higher than bulldozer (they went up to 5 GHz stock ffs) but much less power hungry and all performance gains will come from extra floating point units, shorter pipeline (lesser branch missprediction penalty), sound cache hierarchy, better instruction scheduler and memory controller.