Tuesday, February 16th 2021

Intel Core i9-11900T "Rocket Lake" Processor Allegedly Catches Up with Zen 3 in Single-Threaded Performance

When AMD announced its Ryzen 5000 series of processors based on the new Zen 3 architecture, the performance of these processors was the best on the market. Even in our own testing, we have found that AMD's Zen 3 core is the highest performing core on the market, even beating Intel's latest and greatest, the 10th generation of Core processors. However, Intel has been doing some silent work and the company has developed a new core to be used in the 11th generation "Rocket Lake" platform. Codenamed Cypress Cove, the design is representing a backport of the 10 nm Sunny Cove design, supposed to bring around 19% IPC improvement across the board.

If you were wondering if that was enough to catch up with AMD's Zen 3 IPC performance, look no further because we have Geekbench 5 performance results of Intel's 35 Watt Core i9-11900T processor. Having a base frequency of only 1.51 GHz, the CPU is capable of boosting one or two cores to the very high speed of 4.9 GHz, giving us a good example of the single-threaded performance we can expect from this CPU. In GB5 tests, the Core i9-11900T has managed to score 1717 points in the single-threaded test and 8349 points in multi-threaded results. Comparing that to something like AMD Ryzen 5800X, which scores 1674 points in single-threaded results, Rocket Lake's Cypress Cove core has managed to be 2.5% faster than Zen 3. However, in multi-threaded results, the AMD chip is unmatched as the low TDP of the Intel processor is stopping it from reaching full performance.
Source: Geekbench 5
Add your own comment

90 Comments on Intel Core i9-11900T "Rocket Lake" Processor Allegedly Catches Up with Zen 3 in Single-Threaded Performance

#76
Cobain
evernessinceHardware unboxed already bedunked this claim. They reached out to other retailers and they are not reporting above average RMA rates for Ryzen 5000 series CPUs.

You complained about people bashing Intel and then you went off topic and started bashing AMD. At least the people complaining about Intel are on topic, you are not and are being hypocritical.
AMD unboxed? Who would have thought? :D
thesmokingmanThey deleted their tweet, nuff said!
They deleted the tweet because AMD contacted them to solve the issue. Just like AMD unboxed deleted their tweet bashing NVIDIA 2 months ago, as NVIDIA apologized
Vayra86Haha and here I was just wildly guessing it was nonsense. There we go.
Overclock net forum disagrees. 38% have either doa or bsod from their poll.

Also AMD fórum os full of this same comlain. I linked it.
MakaveliThe whole issue with the original tweet from powergpu before it was deleted. Is that isn't a big enough sample to mean anything if it were true.
They didnt tell About the sample, only odds like "each 100, 4 are faulty" etc

You guys can be on denial if you want. Not long for more retailers to complain about this issue that is mainly affecting new batches. There is even a thread showing the most problemstic ones. But whatever...
Posted on Reply
#77
dyonoctis
CobainAMD unboxed? Who would have thought? :D



They deleted the tweet because AMD contacted them to solve the issue. Just like AMD unboxed deleted their tweet bashing NVIDIA 2 months ago, as NVIDIA apologized


Overclock net forum disagrees. 38% have either doa or bsod from their poll.

Also AMD fórum os full of this same comlain. I linked it.


They didnt tell About the sample, only odds like "each 100, 4 are faulty" etc

You guys can be on denial if you want. Not long for more retailers to complain about this issue that is mainly affecting new batches. There is even a thread showing the most problemstic ones. But whatever...
They actually went as far as deleting their whole twitter account :
twitter.com/PowerGPUcom
Posted on Reply
#78
Max(IT)
MakaveliNo nerve at all boss don't make assumptions.

And I'm well aware of your system specs.

We all need to wait until we have official reviews everyone is interested in seeing Intel's answer to increased competition and more so for Alderlake S. Rocket Lake S is a stop gap solution with no upgrade path and locked to 8 cores.
I have no interest on Rocket Lake.
But I have interest in Intel being competitive, because we saw what happened when AMD took the lead: ridiculous prices for Zen 3.

If they keep fighting for the leadership, we customer will win.
Makavelia 5800X will boost to 4.85Ghz stock not the listed 4.7ghz AMD has on the package.
I can confirm that. My 5800X boost 4850 MHz
Posted on Reply
#79
Makaveli
CobainThey didnt tell About the sample, only odds like "each 100, 4 are faulty" etc

You guys can be on denial if you want. Not long for more retailers to complain about this issue that is mainly affecting new batches. There is even a thread showing the most problemstic ones. But whatever...
I don't think you understand how statistics work.
Posted on Reply
#80
Valantar
londiste@Valantar, Alder Lake should have Golden Cove (Willow Cove successor) cores alongside Gracemont (Tremont successor). By all indications and rumors this is going to be on 10nm.
Beyond that there is Meteor Lake but that is where details get hazy.

Willow Cove has IPC increases over Rocket Lake, primarily due to more cache. Basically the same difference as Ice Lake to Tiger Lake on mobile. Golden Cove reportedly has additional improvements over that.
I guess that's true, but given how minuscule the ICL->TGL IPC increase is, that doesn't bode well overall. TGL gains a few % over ICL, with improved clock speeds representing the vast majority of the performance increase for those chips. So even if Golden Cove is more of an improvement over Sunny Cove (ICL) than Willow Cove (TGL) was, that doesn't really tell us much. Alder Lake will likely beat Zen 3 in terms of IPC, but its clock scaling and power consumption is still an unknown, and even if it's actually made on 10nm (which would mean that Intel has dramatically improved their 10nm yields), that doesn't really point towards that much of an improvement. Peak clocks won't be higher than 14nm parts (likely lower), and efficiency is unlikely to come close to Zen 3 - they're not getting a >50% perf/W increase in a single generation. TGL and the 10SF/10+++ process was a clear improvement over both ICL/10++ and 14nm, but it's not a game changer by any means - even Zen 2 APUs trounce TGL laptops in sustained workloads after all, even if TGL boosts higher and has better ST perf. But there's nothing really indicating that Alder Lake will be a massive improvement for desktop performance, at least over Rocket Lake. It'll no doubt be more efficient, given how clock optimized 14nm+++++ is, its IPC gains are likely to just make it slightly faster overall.
Posted on Reply
#81
Vayra86
CobainAMD unboxed? Who would have thought? :D



They deleted the tweet because AMD contacted them to solve the issue. Just like AMD unboxed deleted their tweet bashing NVIDIA 2 months ago, as NVIDIA apologized


Overclock net forum disagrees. 38% have either doa or bsod from their poll.

Also AMD fórum os full of this same comlain. I linked it.


They didnt tell About the sample, only odds like "each 100, 4 are faulty" etc

You guys can be on denial if you want. Not long for more retailers to complain about this issue that is mainly affecting new batches. There is even a thread showing the most problemstic ones. But whatever...
Fair enough, let's see if the issue sticks then. Bad batches happen. We're also still off-topic ;)
Posted on Reply
#82
Caring1
1d10tWow, there's T now?
Since the early Core Duo T series.
Posted on Reply
#83
DeathtoGnomes
CobainAMD unboxed? Who would have thought? :D



They deleted the tweet because AMD contacted them to solve the issue. Just like AMD unboxed deleted their tweet bashing NVIDIA 2 months ago, as NVIDIA apologized


Overclock net forum disagrees. 38% have either doa or bsod from their poll.

Also AMD fórum os full of this same comlain. I linked it.


They didnt tell About the sample, only odds like "each 100, 4 are faulty" etc

You guys can be on denial if you want. Not long for more retailers to complain about this issue that is mainly affecting new batches. There is even a thread showing the most problemstic ones. But whatever...
I dont see any proof of what you say is true. None. So I'm just gonna call Troll here.
Posted on Reply
#84
thesmokingman
DeathtoGnomesI dont see any proof of what you say is true. None. So I'm just gonna call Troll here.
Smear campaign is easy to spot.
Posted on Reply
#85
Gmr_Chick
Also, PowerGPU's web-site is a fucking joke. I just checked it for shits n' giggles. No type of configurator, nothing to select as a "base model", etc.... Nope. Just a fucking form for you to fill out.

They do, however, tell you up front that the build fee is $150. How considerate. Assholes.
Posted on Reply
#86
1d10t
ValantarAs said above, T-series Intel CPUs have been around for ages. They aren't common among consumers, as ... well, they're limited to a 35W PL2, so they perform a lot worse than no-suffix or K SKUs. This is reflected in their much, much lower base clocks. If you're not able to tell this from a spec sheet, you shouldn't be buying CPUs. They aren't meant for the same markets, but mostly for OEM SFF workstations and similar. (T SKUs are often difficult to find in stock with retailers at all due to very low consumer sales.) Will the 11900T perform better than the cheaper 11700K? NO. And it's not meant to. Outright performance isn't all you can pay for; maximum performance within a small power envelope also has a value (as do many other configurations). So for the 11900T you'll be paying a premium for getting the most powerful 35W Intel CPU around. Which will be slightly faster than the 11700T, which will be slightly faster than the i5-11600T, and so on, and so on. Regular 65W SKUs or 125W K SKUs will beat these every day of the week. And that's fine.
Caring1Since the early Core Duo T series.
I see, its just I've never saw them in retail channel or PIB, let alone geniusbar leak :D
Posted on Reply
#88
Cobain
DeathtoGnomesI dont see any proof of what you say is true. None. So I'm just gonna call Troll here.
I literally linked all sources on my first post

I don't understand though why there is no reaction to the reboots topic with 500+ posts on AMD's official forum.
I don't understand why there is no such topic in Intel's community.
I don't understand why we have 1000+ posts in overclock net topic if things are just fine.
I don't understand why we have 5 different AGESA versions in 3 months if there are no serious issues.
I don't understand how it is possible to get 2 defective CPUs in a row (hard reboots and BSODs on the 1st one, BSODs on the 2nd one)
Posted on Reply
#89
Gmr_Chick
CobainI literally linked all sources on my first post

I don't understand though why there is no reaction to the reboots topic with 500+ posts on AMD's official forum.
I don't understand why there is no such topic in Intel's community.
I don't understand why we have 1000+ posts in overclock net topic if things are just fine.
I don't understand why we have 5 different AGESA versions in 3 months if there are no serious issues.
I don't understand how it is possible to get 2 defective CPUs in a row (hard reboots and BSODs on the 1st one, BSODs on the 2nd one)
Just going to tackle these in order...ahem.

1. Your "sources" are complete bullshit
2. The topic of this thread (way to completely hijack it, btw) is about Intel's Rocket Lake. This topic IS NOT about AMD nor some imaginary defect regarding Zen 3 that you pretty much pulled out of thin air, your ass, or both
3. Again, bullshit
4. OCN is a fucking forum. Just like TPU has a forum. People bitch and moan about problems they more than likely created themselves by tweaking shit they weren't supposed to, or knew the risks and fucking did it anyway because they're a special kind of stupid, and then wonder why their rig doesn't work anymore.
5. Still doesn't mean there are serious issues. People are free to choose to update their BIOS or not. Of course, if it ain't broke, don't fix it. And yet, they "fix" it anyway. Go figure.
6. Is this coming from your PERSONAL experience? Or from that shady-as-fuck "PC Builder" PowerGPU?
Posted on Reply
Add your own comment
Jun 11th, 2024 08:24 EDT change timezone

New Forum Posts

Popular Reviews

Controversial News Posts