No 5Ghz like anticipated

sure, but they're trying to aim their consumer level market directly at gamers. since gamers are mostly retarded and computer illiterate, AMD can impress gamers by showing pretty bar graphs and big numbers.. because bigger numbers always means better or faster for these simple fucks. AMD walks into the room and starts raping virgins for their neet bucks and the gamer faggots will beg for more.

Salty Intel nigger. Not even the most uneducated gaming goober will buy your trash now. Stay salty retard , stay getting btfo in all metrics. Gayms, cinabench, blender and more. Get dabbed on you worthless fool you absolute cretin

Attached: ff.jpg (820x1024, 99K)

Only time will tell.

Bulldozer was seen as a BS lineup for its real world performance even with the clock speed advantage,

If AMD doesn't sell well, I'd say it's more because of brand loyalty to Intel and they'll pick out whatever they can find to use against buying AMD, such as its lower clockspeed.

>*NEW* ABSOLUTELY NOTHING MATTERS AT ALL! REEEEEEE

intcel cope
have sex

The big question is *why* they don't reach 5 GHz. If AMD could release a 5 GHz part I'm sure they would, if only for the marketing value. Is Intel's 14 nm process just simply better than even TSMC's 7 nm in terms of switching performance, or is Zen simply not as well optimized for cycle time as Core is?

Keep in mind that Core has been overclockable to 5 GHz since like Sandy Bridge. Though I guess it remains to be seen how overclockanle Zen 2 actually is.

>muh nigguhertz
You're the fucktard who would buy a 3.6 GHz Pentium D over the 2.4GHz Core 2 Duo just because MUH Clock speeds.

>AMDtard coping that nigguhertz doesn't matter
>when a day before this same AMDtard is shilling for a 5Ghz Ryzen
KEK

this
ryzen has better IPC than intel right now and if someone managed to get his 3900x up to 5ghz it would annihilate everything intel has to offer

>Though I guess it remains to be seen how overclockanle Zen 2 actually is.

With the 12 core doing 4.6ghz, I see no reason why the 6 and 8 cores can't be OC'd to that level as well.
Maybe we'll see golden chips hitting a 4.7-4.8 come July. That'll be exciting to see.

ryzen doesnt need 5ghz to beat intel housefire garbage

Except that wasn't the point, idiot. If Frequency alone is important, we would be having 10GHz single core monsters, and the Pentium brand would still be the premium intel brand.

Learn your history, kid. Most of the technologies that we are taking advantage of today:

> Multi-core CPU's
> on-die memory controllers
> X64

are all AMD-pioneered. While Intel was just going

>MUH NIGGUHERTZ

man that cant be a dude.

5 years ago the hot shit for intcels was
>MUHHH IPC CLOCK SPEED DOESNT MATTER
and nowdays its the opposite

14nm isn't, but the selectively binned top 5% of the 14nm++++ refresh is.

see what it gets at 1.40V

Attached: 1457199777628.jpg (319x310, 43K)

>Can't refute that Intel adopted the same approach, when they forged the original Core architecture.

This is how you see that the inturd shill is really just that. A shill, it's a shame that he doesn't know his own company's history.

Genuine question, with the nonexistant TDP of the new chips (especially 3700X), what's preventing higher clocks? I thought the main limit was heat, which is directly proportional to power consumption. So why can't you take the new 65W chip and raise the clocks until it eats like 95-105W and have much better marketing value?

Nothing, it already beats the 9900k for lower power usage.
Plus there's one that's 9900k equivalent for cheaper and lower power too.

Attached: Laughing Whore.jpg (762x900, 161K)

Maybe, just maybe, the ones saying that are the same people who are brand neutral and appreciate good engineering.

Confirmed boost clocks are easily achievable on any silicon for 24/7 use.
But it's rendered dumb to manually overclock thanks to PBO.

REMINDER THAT OFFICIAL SCORES AND BENCHMARKS DON'T USE PBO.

By your logic, Brodozer was superior to Sandy Vagina.
Long live Coffeelake.

>coffee lake
refreshed garbage
long live zen 2

My guess is that the 3800x is exactly what you're describing whereas the 3700x is their way of showing how far ahead they are in perf/w.

Way to completely misinterpret the post

If that's the case then it sounds like you can save $70 by overclocking manually. That said, the 3800X is only a little bit higher clocked - 300Mhz higher base, literally 100MHz higher boost - and that's a jump from 65W to 105W; is that really right?

If the Intel Shill was in charge of Intel during the 2000's we probably would've seen Penitum V for 5GHz Single-Core with HT against the Athlon 64, and we would've seen literal housefires

>what's preventing higher clocks? I thought the main limit was heat, which is directly proportional to power consumption.
Not really. With intels "refined process" they have basically aimed to make the CPU stable at higher clocks. Sure - the 9900k will use 250W but it can keep stable with a fuckhuge cooler.

You can't do that with zen1 (and zen+). Once you go over 4.0ghz on zen1 the power consumption goes up drastically, but it crashes way before you can get to 250w. This is because their process was designed for low power. Looks like the new zen2 is on low power too. They make up for it with IPC though.

>anticipated
by whom? did amd ever hinted about a turbo clock of 5ghz? no
the only thing we might expect is the 16c at 5ghz

That seems right, yeah. 3800x would probably be a higher binned 8-core chiplet but you can reach the same performance with a 3700x at a higher voltage and power usage.

The 3800X is the cpu that will be able to hit 5ghz hence the 105w tdp rating
the 3700X will still be able to beat the i9 9900k at 4.5ghz but the 3800x will beat it by a long shot

>3800X is the cpu that will be able to hit 5ghz hence the 105w tdp rating
Let's not pull info out of our asses here or there'll only be disappointment. If it could hit 5ghz, I have no doubt that AMD would have rated their chips as such.
But we'll only find out in July when people start testing retail chips.

BULLSHIT

Higher speed is always better. Throwing more cores onto a die that never get used does not benefit anyone.

There is a reason no one is using Ryzen for laptops. Core 0 speed is king and I didn't need to do a comp sci degree to know that but it helped

Lol ok. Enjoy your 50% higher power usage for the last 3% of performance.

Good point
the 15% ipc increase and the 500mhz increase is still great for zen 2 and im expecting all the X SKUs from ryzen 4000 to hit 5ghz with no problems

>4.6Ghz
This is good enough to destroy Intels 5.5Ghz, next get CPU will probably get to 5Ghz and destroy intels hope for a very long time.

I am only tempted to upgrade to this as I think board manufacturers might pull some bs about supporting next years chips like MSI tried to get away with this year.

Also im not 100% sure if this is the last series for AM4. The "support up to 2020" is so vague.

Have sex

ok then not 5ghz but 4.7 or 4.8ghz with a manual overclock on the 3800x could be possible
besides that why would amd give the 3800x a 105W tdp if it only has a 300mhz base clock increase?

seething

not the same person

You won't have higher power usage if there are less cores to power.

Did AMD withhold 5ghz parts in the wake of Intel losing a bunch of performance with the recent security flaw?

Is AMD sandbagging and intend to release it later because now they're -so- far ahead?

Attached: 1554440738663.jpg (804x802, 85K)

>Higher speed is always better. Throwing more cores onto a die that never get used does not benefit anyone.

Then why would Intel ditch the higher-clocked, single core with HT Pentium 4 in favor of the lower-clocked, dual-core Core 2?

AMD is indeed sandbagging
they will release a 16 core part but they didnt show it in the keynote

Why would intel suddenly stop making quad cores and move to making 6 cores and 8 cores?

>6C 4.4
>8C 4.5
>12C 4.6
Smart money's on "yes".

They just want to curbstomp right when they announce their answer to the current lineup.

Also it may be due to the commitment to support AM4.

6 cores? What do you need 4 cores for? 2 cores are ought to be enough for anybody.

It was already known that 16c wouldn't be announced now.
No reason to. They'll wait for Intel's next move (10c aberration with a million tdp).

>300Mhz higher base, literally 100MHz higher boost - and that's a jump from 65W to 105W; is that really right?
Boost doesn't have to mean the same thing. That TDP headroom could help it boost more with more cores.
3700x could be all core 3.6ghz, 4cores 3.8ghz, 1core 4.4ghz
3800x could be all core 3.9-4.1ghz, 4cores 4.3ghz, 1core 4.5ghz

>If that's the case then it sounds like you can save $70 by overclocking manually.
>3800x would probably be a higher binned 8-core chiplet but you can reach the same performance with a 3700x
If it's anything like zen+ then manual OC will not work great. Remember, manual OC didn't reach the boost speed of STOCK 2700x. PBO... Maybe?

nah 1 core is all you need fellow intelbro

>Is AMD sandbagging
The cunts had better not be, it's such a douche move
If there's a 16 core that gets to 4.8 I'm in

>but the selectively binned top 5% of the 14nm++++ refresh is
Sure, but I don't think there has been so much as one single sample of Zen 1 that has been overclockable to 5 GHz.

Its not a douche move, its a legitmate strategy to bait the competition
and in this case its intel

It'll do consistent 144fps in Fortnite like zoomers want. Even Zen+ Threadripper/2700x already demolished intel in multithreaded performance anyway. So now AMD has
>single threaded parity with Intel
>MASSIVELY outperforms Intel on multithreaded perf
>MASSIVELY outperforms Intel on price/performance
Buying intel will be MAYBE worth it only for people with 240Hz 1080p monitors, i.e. less that 0.1% of users. Even then, the top 3900X chip might outperform the 9900KS on single threaded performance considering the 15% IPC uplift.

Anyone who buys Intel for desktop for the next 2-3 years will have made a retarded decision.

hello, china. stop eating dogs and stop supporting communism.

My core2duo clocked to 3.6ghz with the stock fan, so did my core2quad. They had the market sewn up so they released processors at different speeds to make more money. Prescott was older architecture and at its limit.

Faster single core performance is why Intel are still selling k procs at stupidly profitable prices

the 2950x and 2990wx can easily get to 4.3 and 4.4ghz
but they obviously are golden samples and they sell for nearly 2 thousand dollars
im sure they will do the same with the 12 core and 16 core part once they get released

So I guess you, then, are the retard who would've bought Iranium over any other contemporary processor whatsoever because MUH IPC.

>If Frequency alone is important
Noone pretended that frequency alone is important, but that doesn't mean that frequency doesn't matter.

lmao

Attached: Screenshot_20190527-135257~01.png (1080x400, 223K)

I have a 20 core Intel cpu in my server at work since 2016. It's not going in my desktop or laptop anytime soon even if it was cheap because it's fastest core is still too slow. Same with the new Ryzen. Until I see.some over clocking benchmarks from independent reviews I'm not shelling out

that's a wolfdale core, they are a better refined Core 2. The early Core2's were not as high-clocked as their Netburst counterparts.

>Anyone who buys Intel for desktop for the next 2-3 years will have made a retarded decision.
> a decision based on processors that don't exist yet
> ... based on really sketchy figures of CPUs coming to market, figures pulled out of AMD's prolapsing communist asshole
AMDrone shitposting: Jow ForumsENIUS LEVEL

>parity with 9900k
>comparing this to a 3.4ghz server CPU

Kike

>if they are, you're buying reject bins
>if they aren't, you're buying early as fuck unrefined shit

Attached: 1446786993767.jpg (706x706, 73K)

what will intel release in the next 2 years? copelake with 2 more cores each generation and with no IPC improvements?

>t-they're a bunch of commies!
Is this how intel is coping?

>muh nigguhertz
Except that wasn't at all what I was saying, retard. I was merely posing the question why they are inferior to Intel in that particular metric, not putting that metric as superior to other metrics.

How long before Intel goes crying to papa Trump and Papa Trump Huawei's AMD.

Intcels are on suicide watch

The difference of 4.4ghz at 65w to 4.6ghz at 105w makes me thing there's a scaling wall again. I never expected going above 4.5ghz to be easy but if the single core boost doesn't go above that then I suspect it doesn't go much higher. There's a chance it's just heat and power draw that can be overclocked through with appropriate cooling but there's still a good chance it's process again and everything becomes unstable above that without exotic solutions.
Remember even intel's 14nm++++++ is pretty tame up to 4.5ghz.

7nm will probably come from of the new oregon plant, which won't be ready for a couple years

AMD is an american company

That metric wouldn't make sense without context. So again, In your little mind. The P4 is superior to the Core 2 because it can clock higher.

Go away RTG, start binning your parts properly.

wow. amd caught up with single core performance levels.. how many decades did that take those useless asians? three decades? lmao.

by the time intel makes a 7nm cpu AMD will have already moved to 3nm

I will, all I care about is rendering/compiling/simulation performance. Couldn't give less of a shit about 20% single threaded performance one way or the other, if it scores higher in Cinebench and Corona I'm good.

Not really. Again, learn history, shill.

see
>That TDP headroom could help it boost more with more cores.

for an american company, it sure has a lot of communists from china working in their US base.

Not entirely correct I had both the 65nm and 45nm core2 duo and quad. The duo was 1.8ghz over clocked to 3.6 and cost 115euro with a fan, the motherboard was 65 euro and is compatible with both of procs. Now that was value. Also the reason they clocked so high was because the memory controller was off chip (northbridge) which AMD fans laud as their amazing innovation. Not so great for max performance though, as first gen i7 can attest to

Intel 10nm isn't coming to desktop until AT LEAST late 2020-early 2021. Zen2 is about to absolutely demolish Intel's current offerings, and we already know they cant retaliate by either increasing IPC (in fact it's decreasing due to hardware vulnerabilities) or clocks (9900k is already a housefire). They WILL release a chip that'll do marginally better than the 3900X but it'll almost certainly be more expensive than even the 9900k because it will have to be golden sample chips.

Meanwhile, AMD is moving to TSMC 7nm EUV that'll likely provide improvements similar to Zen+->Zen2.

Intel is an american company and most of its workers are from india
is this the reason intel stock coolers smell like curry?

>So again, In your little mind. The P4 is superior to the Core 2 because it can clock higher.
So again, I didn't say that frequency alone made it superior. It still doesn't answer the question why Core clocks higher. As I said, it would be interesting to know if it's a process issue, or an architecture issue.

and intel caught up with multithreaded performance
how cute

Wouldn't 7nm -> 7nm+ be more in line with 14nm -> 12nm? I know they're calling it Zen 3 instead of Zen2+, but still.