New Generation Intel CPU's 'Tiger Lake' Processors

Both wrong, sorry. They aren't equivalent to the Intel-based platforms. Nobody I know kept the AMD version. Nobody.

I am not talking about the CPU (only). It's the platform that is the special sauce...

And we were talking Surface, not ThinkPad's.

The 15" model is below:

See:
https://www.microsoft.com/en-us/sto...=&selectedColor=CBB1A0&preview=&previewModes=

 

@tilleroftheearth don't you feel betrayed by Intel with their security screw-ups? I personally do. They knowingly pushed vulnerable hardware that is impossible to fix. CSME vulnerability was last straw for me. I don't know when will I be able to trust Intel as a CPU vendor again. AMD is the only reasonable alternative, at least at the moment.

 

Lol, ok so no one ever keeps their AMD thinkpads because it has a different CPU, good to know. I guess the 3 that I have had with AMD cpus were probably just mislabeled because they worked great without issue. I hate when they put the wrong brand sticker on the laptop, it sure fooled me. Again, you think I'm wrong and I believe that your making unrealistic statements. The beauty of having opinions I suppose.

 

There must have been demand for it, otherwise Microsoft would not have included it in their product line. Of course, you think that because something doesn't meet your standards, it's not good enough for literally anyone else. Your history of posting about SSDs (everyone needs minimum 4 TB SSDs provisioned to 85% or else you can't do any "real work") is further proof.

Define "platform." Be specific.

 

Between the AMD and Intel (Ice Lake) Versions for Surface I’d have to agree the Intel with one is better, if I recall from past tests ranging from battery life to synthetic, it was better.

For Desktop I’d lean towards AMD as Intel is still stuck using Skylake era cores at 14nm and guzzles more power to reach equivalent performance. I will say that is for my workloads, before anyone points some niche things that one or the other is better at.. and for some of my work those extra cores definately do scale well.

Downside to AMD’s 30+ Watt mobile chips while I believe they are superior to Cannonlake is that options are fewer, as OEMs don’t always pair them with the highest end GPUs and at times give them shoddy cooling compared to equivalent Intel chips, one laptop for example, don’t recall if it was ASUS or Lenovo, or maybe it was Dell had their AMD version with less heat pipes and a single cooler while the Intel version had a dual cooler.

 

I'm going to copy/paste a reddit post that offers a good explanation as to why this is:

Source

I would like to see which one it is you're talking about. However, laptops like the refreshed HP Omen 15 and Lenovo Legion 5/5i, have identical cooling setups between the Intel and AMD models.

 

@Starlight5, I don't care about any security screwups from Intel. My workflows account for worse. Even better now that they are 'out'. Anyone using an AMD platform should be concerned, IMO. If any vulnerabilities get exposed, I'm sure they'll move slower than Intel was able to. Thinking you're safe and being safe isn't the same thing. Silicon is silicon. There is nothing bullet-proof. Not even from AMD.

@custom90gt, quoting myself "No one I know...". If I knew you, I would have said 'except for three...'. Happy now? These are not 'unrealistic' statements. Your assumptions are. My statements and opinions are based on facts.

@saturnotaku, like @custom90gt, your assumptions are false. If one wants a storage subsystem that can deliver as close to advertised specs as possible, no matter how it is used, over time, while also maximizing the life of the nand, then buy the highest quality, biggest capacity SSD and overprovision it by 25% or more (and just as important; max out the RAM while you're at it too - 32GB is the new minimum). My testing shows that at around 33% the loss of additional capacity isn't equal to the increase in sustained performance over time, nor does it significantly increase the nand's lifespan either (this is for an SSD that is used as a system/boot drive. For SSD's that I use simply as Scratch Disks/Temp spaces that get hammered 24/7, I still OP to 65% or more, depending on how long I find the SSD to last and at what price(s) I can source them for). There, I corrected you on that too. I do not assume what people want. I tell them what is achievable and let them decide if they want that experience for themselves or not too.

As for the AMD Surface? Bad choices by Microsoft. Even with their considerable expertise, it was still a pig with lipstick when productivity is the name of the game.

The platform, how I was speaking of it before, is closest to Intel's Evo, aka Athena (2). Apple, Microsoft, Lenovo (ThinkPad workstations, specifically) and Intel only hold that distinction so far (and not for all products, of course). Not that the mentioned companies are (all) in the Evo group. Rather, they each have/had the balls to go off in a different direction and stick to it until it was fruitful. With at least one product or another.

The interactions/meshing between man and machine is infinitely more important than the machine itself. That kind of platform.

@Aivxtla, the generation of the platform (note: different platform meant here) doesn't stop me from considering using it. The performance, stability, reliability, and compatibility are all equally important when time is $$$.

 

@saturnotaku I think it may be the Dell G5 15 SE, hopefully I have the right one. I feel like I should just have bookmarks for times like this lol.
https://www.notebookcheck.net/Dell-...Ryzen-4000-gaming-laptop-option.469723.0.html

 

Sorry @tilleroftheearth, I haven't seen any facts from you, only opinions. I'll leave it there. We won't see eye to eye, and I'll continue to run whatever is on top, which I believe is AMD at this point (from reviewers and my own testing). We will see how well they do with 11th gen, but I wasn't very impressed with my 1065g7 Ice lake... It would be nice if Intel would commit to PCIE 4.0, they seem to be wishy washy on this.

 

That's... confusing. Even if CPU vulnerabilites are not in your threat model, performance hit Intel took from mitigations is quite noticeable. Nevertheless, I guess we'll leave it at that.

 

@custom90gt, I'm sorry too, because the facts are right in front of your eyes. Reviewers don't tell me what works for me. Go read some of my first posts here, I didn't care about 'scores' back then and I still have no reason to believe them now. For my workflows. I wasn't too impressed with 1065G7 Ice Lake either. Hence, I don't own one (even though I tested a few).

Intel has not only committed to PCIe 4.0 on mobile (which is where things seem to get introduced from them), they have already planned and executed DDR5 support too with Tiger Lake. Are the reviewers you believe still laughing at that? Yeah, I'm sure there will be some caveats, somehow. But all indications are it will arrive before the holidays this year.

@Starlight5, the performance hit is just a 'market correction' that affects everything/everyone, it's a non-issue where security is concerned. When Bezos is worth $200B and the market hiccups, he isn't worth any more or any less than the day before compared to his peers. It's relative. I'm not worried about it. I'm not playing games with my tools, I'm earning $$$ from them. Do you know how I make up for a 20% hit? I buy more workstations and hire more people. See? Just numbers, just the cost of doing business. Guess who pays for that? Not me. My entire client base does (just a little extra spread out over them more than makes up for any hiccups Intel has had to endure from enterprising hackers that have affected me, directly).

Not that I'm in Bezos' league. But that is a fine example of why Intel's 'woes' aren't what they're reported to be on online rags and video blogs for most who use their tools for productivity, instead of entertainment/hobby.

 

All of which--if I'm not mistaken--is simply to restate that:

There are consumers/prosumers: enthusiasts, hobbyists, gamers, OCers, average joes.

Then there are commercial customers.

Priorities and opinions vary widely across the rift which separates those who use computers for business, and those for whom the use of computers is reducible to a function of their business' income.

I believe this disconnect is of deep, even philosophical nature. If you are forced to discuss tools and equipment as a relation of cashflow, the perspective naturally differs from that of a tool that only needs to exist in the role assigned to it and enable your operational structure.

When the metric is time=money and tool=time, there is zero room for brand loyalty or consideration to rhetoricals.

 

@tilleroftheearth If you wouldn't mind posting some links, that would be appreciated by everyone. You have more than 12,000 posts.
There is frustration in this thread over what you have been saying because, so far, you have not backed up your statements with anything objective.

Charles

 

@Charles P. Jefferies, thank you. But it is just as hard for me to find anything from the first few days/weeks/months I joined too. I just thought an admin/moderator would be able to access more powerful search criteria than a user can.

 

My search controls are the same as yours. It may be easier to search this forum using Google: site:forum.notebookreview.com my search terms.

I would like to see the posts if you can find them. Otherwise why not type them up again ...

Charles

 

Retype them from 2009/2010? Lol...

Which points do you want to see, exactly?

 

Well, in my area the ambient temps are above 35C so I'm literally running out of spec.

I think tillerofearth requires vPro IT firmware access/privilege via OS deployment packages that enterprise use to deploy OS, apps via some provisioning which might be incompatible with AMD. Also, enterprise don't always use latest SW versions for their multi-million dollars SW licenses say CATIA, Autodesk Maya, Civil 3d etc... where on Intel its just deploy and all works fine w/o needing major recompiled code for future Intel uArchs unlike AMD which might prefer generalized Compiler like gcc, MSFT to make programs run at same speed on AMD, Intel, Arm64 etc...

 

Can't you answer those questions on your own?

Charles

 

I can and I did, to the best of my ability.

I can't know how to answer for others lack of understanding though. The crystal ball gets a little hazy there.

 

Your facts really just consist of your opinions, I haven't seen anything factual from you. That's ok though, you don't have to go digging through your old posts just to find some facts for me. Yes you hate AMD, yes you have 3 friends with 3 Lenovo laptops that had AMD cpus in it and none kept them, no sense in beating that dead horse. We will see what the new CPUs bring, but I'm skeptical until I see something from a trustworthy source (and I don't believe Intel making up a new term called RUGS and then giving me some arbitrary numbers on a couple of products I've never heard of, or things I don't care about such as 1.2x faster in word, is a trustworthy source). Again, you have your opinions and I have mine. Simply stating that your opinions are facts doesn't make them so...

 

But I was so excited about intel's numbers, especially the faster artificial intelligence, because everyone was getting laptops with the 4800U to do that and this new generation of cpus is a total game changer

 

@custom90gt, I haven't seen anything factual from you either. Yet I still respect your opinion and try to reply to your posts as fully as I can. All you do is ignore anything you don't want to hear/read. No problem.

You also like to try to make me look as stupid as possible, that's all right too because that strategy usually backfires. But this certainly isn't becoming of a Super Moderator.

I like to see things with an open mind and my conclusions are fact-based. Sorry, they're so harsh for your 'team'.

For the record, I have posted in this thread the video where 'RUGS' is explained, I posted the link where the benchmarks and their setups were detailed. I answered to the best of my ability your direct questions. But, of course, you're entitled to ignore it all. My past remarks are not needed here today. I was treated the same back then, thank you.

Nobody is making you care how much faster Word is, or the other benchmarks that show the disparities between synthetic benchmarks and real-world workloads. The point is I do. Because that is 'real'. If you don't have anything further to add to this, please stop attacking me and my posts.

 

I fully admitted that everything I was saying was an opinion, I never tried to push them as facts. I didn't ignore your posting, but when you post something like "the fact is intel offers a better user experience" without any actual facts backing it (outside of an Intel sponsored video that says it's better in the random benchmarks that it happens to be better in), I question it. Intel is hand picking synthetic benchmarks and then spinning that they are more important than things like Cinebench, probably because they are benchmarks that they are actually faster in. If running Word at 1.4x means something to you, than great. I haven't a clue what that really means since I have ran Word several hundred different CPUs without noticing any real difference.

I don't try to make you look "as stupid as possible," I am simply questioning your posts when you keep telling me you have facts. Sorry if you feel like I am attacking you, that's not my goal. My goal is to try to figure out what your real hang ups are on AMD. I am not attacking you or your posts, but questioning them. Yes even moderators are allowed to have opinions and question the validity of things like "Smarter cores are simply better for real-world workflows that most people use." And I do believe you initially started the conversation after I posted a video showing the 4800u being 2x as fast in what I would consider real world testing, or my own RUGS per say.

 

I can accept your response @custom90gt, and I appreciate it.

I am still not positive that you're understanding what I am saying. Sure, it's a PR video. Sure, it's 'evil' Intel telling us more lies. I get that. But I'll ask again. Have you ever used (fully) any of the Surface products for an extended period? I'll repeat that I had tested them and returned them more times than most, but at some point, the idea, the hardware and the O/S and software gelled to what it is today and it is world-class, for a reason. Even if I can't put most of those reasons in words people here can understand easily without also experiencing it the same way.

I happen to believe that Athena, Athena 2 (Intel Evo), and similar forward-thinking collaborations between normally disjointed aspects of tech is what really is pushing the envelope further. Rather than just a fire breathing, many-core CPU that AMD is pushing hard, Intel is caring (and has for a long time) for the whole package and the entire experience. Much like Microsoft's Surface lineup do too.

This isn't just a belief. It is implemented and demonstrated each time I use such products.

I am not a fan of a company. I am a fan of executed perfection (or the best example thereof). And of course, for my specific uses. That includes never having 'gamed' on a PC, or really getting into overclocking/building (though I did both, superficially).

What excites me about technology is not the underlying tech, but the implementation of that tech down to the smallest details towards solving my problems/workflows. Productivity. Intel, Microsoft, and ThinkPad's are the products (and services) that do that at the highest level currently. This doesn't blind me for all things Intel, all things MS, or all things Lenovo.

I constantly sample, test, and verify that what I have is at the pinnacle of my expectations. Those main players keep coming up winners for me. But where I was once an IBM supporter/user, I discovered Lenovo (no choice, of course). Where ThinkPad's left me wanting more, Surface has proven the superior tool, and so it goes.

And I'm guessing that soon, there will be an Intel Evo based notebook that will challenge and even take over the roles of the devices I use now. Even if they aren't Lenovo/ThinkPad or Surface devices, I will test them for the 'ether' that I can't seem to explain to you. This isn't 'scores' in benchmarks. This isn't dazzling new Nx node tech. Nor is it how many cores per $$$ I can buy. It is much simpler and direct. It is the tool, the model, the company that delivers the complete setup that gets me home to my family faster. Everything else is just so many unicorns and fairy dust to me.

The tech required to safely get men on the moon and back to earth in 1969 is laughable today. It was accomplished with teamwork and a single vision. Today, with all the tech available, that hasn't been equaled, still.

Vision is more important than hardware.

AMD has (some) great hardware right now. Intel has that and a strategy too for theirs.

Go back and look at the video and see the disconnect between synthetic vs. real-world workflows (the same points I made in 2009...).

I don't run benchmarks with my computers to make money with. I use computers closer to how Intel is suggesting.

Hope you have a great weekend.

 

See:
https://www.tomshardware.com/news/amd-vs-intel-which-pc-boots-faster

Here is another small example of where real-world trumps pure horsepower. Let's not get too deep into this (specific article) ...

It's just meant to illustrate what I'm talking about AMD vs. Intel platforms, overall, for my usage.

Raw HP, raw talent, raw potential means little, on its own. Expand the testing in the link above to the entire workday/week/month/year (complete) cycle and you'll get a hint of why I'm not an AMD believer, yet. No hang-ups with AMD, really. Just refuse to go below the level of productivity I'm already at.

 

And if we knew more precisely how you used them, we could better understand your perspective. However, the fact that you continuously talk in vague statements about your own work while treating Intel marketing material as gospel, then act like a victim when questioned, makes you sound like nothing more than a used car salesman.

My work MacBook Pro, and my child's school Chromebook start up, shut down, and reboot faster than any of my "normal" desktop or laptop computers. Maybe you should switch to them yourself if that's a metric you go by.

 

What is there not to understand in my quoted text in your post above? Or, the post above yours? Nuances are vague only to those who can't see them.

Did you see the part about not 'getting too deep' into this specific article? The startup, shutdown speeds are not my main workflow either... yet they are beneficial when opening a notebook on the road two hundred times a day.

I am not a victim and why are you devaluing used car salesmen?

Look at the information I have already posted. If you can't make heads or tails out of it, not my loss, sincerely sorry.

 

I have not used any surfaces just because it really doesn't interest me in any particular way. I've used some 2-in-1s such as my last dell XPS 13" and it was fine, but I didn't utilize it much. I do get the point of Intel working with laptops, I used to attend their events when they were pushing Centrino as we were an Intel Gold Seller and they were always looking at making their products more power efficient and fluid with the rest of the hardware. Having said that, I haven't noticed any usability differences in all of my AMD laptops when compared to all of my Intel laptops (many thinkpads of both variants included). I haven't had any hiccups outside of some of the silly things Dell did with their XPS 15" laptops...

 

YOUR usage.
I noticed in most of your posts when recommending hw to others that you essentially DOUBLE the specs you recommend to others (in terms of SSD storage, you recommend 4TB rather than 2TB, or 64GB of RAM instead of 32GB... claiming that the lower option is also outdated in the process).
Bear in mind that most people do not have the means to AFFORD those recommendations... we need to be mindful of how we invest and where.
If that means going for 1TB or 2TB SSD and 32GB RAM because its most affordable, then we WILL go for it... and also, it doesn't mean those specs are outdated (actually far from it... those kinds of specs would be more than enough for people even doing productivity workloads.... such as myself - doubling the RAM to 64GB would be justifiable when working with enormous 3d scenes for example or processing humongous videos - useful? Yes, the more RAM the better, but the price also usually doubles which is not affordable for most people - and also, you need to check whether your usage scenario will actually EXCEED 32GB RAM or not... if not, then you're effectively wasting money).

Do I think that those double specs SHOULD be STANDARD today and offered to EVERYONE at affordable price? Of course I do, but you should know by now that the market can be ridiculously SLOW in adopting these things and even less so in making them AFFORDABLE to people.
SSD's were out on the market for a VERY long time and they are still ridiculously expensive compared to high capacity HDD's (even though they should have overtaken HDD's in affordability by now).

Also, let's take an example from Intel's page. They kept the market for a LONG period of time at 4 cores and 8 threads.
They didn't really make any decisive push in the consumer market, and also, software developers became complacent as well because multi-threaded games and general software were supposed to be able to take full advantage of the hw one has for a while now, and we're still not there yet.

Even those of us who know how to save money won't blow all of our savings on a laptop with all the bells and whistles when we know that something a lot cheaper can do us just fine for years to come and that we could over time invest a bit further into the system to upgrade its storage and RAM perhaps.

8GB RAM was seen as an 'more than enough' back in 2010... whereas 16GB should have been more or less a standard by 2012 (and yet its only now [2020] that most laptops are coming with 16GB RAM - heck in various cases they still come with only 8GB - and OEM's STILL heavily overcharge on adding more RAM or some don't give the option to upgrade at all because they soldered the RAM sticks to the mobo).

Honestly, there is nothing 'raw' about Intel (except perhaps their greed and propensity to bribe OEM's to keep their products afloat - both of which they demonstrated on ample occasions).

Fact of the matter is, AMD takes a bit longer to POST... simple as that (in reality, if you take away the POST delay, everything else is pretty much on par since booting etc. relies predominantly on the storage drive speed).
That's basically the reason why Intel boots faster.

One of the possible explanations as to WHY AMD is posting longer than Intel could be in the fact that AMD includes microcode support for multiple CPU generations into a BIOS, whereas Intel does not.
AMD is also a lot more security focused and includes those features on a hw level for the most part, and frequently includes them on a BIOS level to patch things up if new threats emerge that need to be patched up - which wasn't much of a problem for AMD (Intel on the other hand ended up numerous security issues over the last few years).

Also, I'm not overtly acknowledging about MS Office 'benchmarks' as Intel uses them... because, using the said software on AMD 'feels' just as fast and responsive... and I really don't understand just what is Intel aiming at when using MS Office benchmarks and claiming they are '25%-80% faster'.
Honestly, opening and saving documents, or the system processing them is pretty much instant for me on my AMD system (and I use Libre Office - though majority of software on my laptop doesn't even use it to its fullest... except 3d Studio Max and Blender - sure, more performance doesn't hurt, but Intel is effectively 'brute forcing' their way to 'win benchmarks' and using dubious ones at best where their hw shines most which is ironic, because Intel also said that 'benchmarks' [Cynebench] aren't somehow an accurate level of measuring real life performance (its most certainly relevant to me and many content creators because those kinds of workloads tend to stress your hw to the maximum).

Intel booting or restarting faster than AMD is hardly relevant. Most people won't notice the difference in question, and to be fair this will vary from system to system.

 

If memory serves me right, majority of the softwares were tuned to Intel performance. It's only after the end of Ryzen 2000series, we see gleams of light shining on the path towards AMD that numbers of softwares were tuned to "blend" with AMD.
There are still numbers of game titles as well as other softwares "on the Intel path" waiting to be guide to run on AMD track...

 

@Deks, what specs am I doubling to others vs. whom else? Give me a link if you want an answer.

The rest of your post shows how little you're paying attention to the posts above yours.

And splitting up the quote to belittle a different definition of 'raw' is not doing you any favors for whatever argument you may have.

My 'usage' is not 'X' or 'Y' or 'X and Y'. My usage is the sum of touching my devices from the first minute of each day to the last minute I interact with them and it's time to go home. The total package. The whole enchilada. The entire process from start to finish each day, each week, each month, and repeated endlessly each year. That is my benchmark.

What you seem to be missing is that 4C/8T (or the Intel versions of higher core count platforms) is still the standard of many, many workflows. Including mine for a variety of sub-flows.

What you need to understand is not just that MS Office is faster. MS Office is faster than on an otherwise more beefed up platform. Nuances matter. Even if you don't concede or 'feel' them in your workflows.

There is no brute force used here by Intel. Education is worth what the learner puts into it. (Synthetic) benchmarks are not the answer to who 'wins' in tech. Actual workflows being completed faster, are.

 

Intel confirms Tiger-Lake H is with up to coming 8 cores. I guess this shouldn't surprise anyone that Tiger Lake-H was going to happen, but good to know it's confirmed either way.
There also seem to be 35w and 45w designs, with the former focusing more on the iGPU (but tops out at 4 cores) than the latter.

 

This too.
A lot of software out there is developed with Intel in mind (and NV Cuda), not AMD and its open source features.
Recent pushes have been seen to optimize some software for Ryzen (or more to the point, multi-core performance in case of Adobe - and Adobe has been including LIMITED support at best - they still hadn't included Vega igp's into support despite the fact its FAR more powerful than all Intel iGP's presently available on the market)..

So, Intel still has a bit of an 'advantage' because of coding being in their favor... but the fact AMD plainly OUTSHINES Intel in both IPC and multicore performance speaks volumes. Imagine if software is fully optimized to make use of Ryzen uArch and AMD gpu's?
AMD had massive compute performance at its disposal in GCN uArch (Polaris and Vega) that went largely unused in games for example and even some professional software because devs were mostly paid by Nvidia to make use of CUDA (which intricately speaking has 0 advantage compared to OpenCL which can run across all hw and performance of which can be further optimized through drivers)... and we've seen just how fast Navi GPU's are when games and software are properly optimized for them as well.

 

Yeah, it will thighten the pressure on AMD. Not so sure AMD is ready for more cores for 35w and 45w notebook designs.

 

OUTSHINES is a little over the top, but yeah, nobody is doubting the hardware was more powerful than what Intel had 3 years ago. Getting work done is rarely about using the 'best' for a single metric (hardware), rather, the 'best' overall.

I can't (just) imagine software fully optimized for use with Ryzen uArch... I need it implemented and available yesterday for it to be worth talking about.

See:
https://www.tomshardware.com/news/intel-makes-it-official-eight-core-tiger-lake-chips-coming

Still waiting for a balanced review of Tiger Lake (7-28W) chips, but these upcoming Tiger Lake 45W/65W chips aren't being introduced arbitrarily by Intel either. Can't wait to see a proper workstation review of both AMD and Intel facing off with their respective best.

 

isn't the current high performance H-class CPUs are 45w makes?

 

Yeah, with Configurable TDP-up to 65 W at least for intel's BGA H-class Top dog.

 

Speaking of the current BGA H-class Top Dog... it looks like demand exceeded supply on 10875H just like the 10900K, and Intel ran out so quietly released 10870H with -100Mhz base clock as a stopgap

 

Intel have finally added in Recommended Customer Price.


They milk (bin) the silicon for every drop nowadays. Bin it as low tier i7-10750H won't please the shareholders. And many of the notebook buyers out there won't see the performance difference anyway As long its labeled with the i7 tag, all will happily buy it.

"Compared to the 10875H, the 10870H offers a 100 MHz lower base clock. This does not sound like a huge difference, but it might be something worth considering as many enthusiasts are often encouraged to disable Intel Turbo. This is due to the fact that turbo does not provide significant performance improvement, but it comes at the cost of much higher temperatures".

Encouraged to disable Intel Turbo? Nope, rather forced due awful/terrible and cheaply engineered cooling!!

 

If I'm not wrong, the Klockable CPUs are the "K"-class for Intel, while Ryzen CPUs are mostly "locked" until the intrusion of external software?

 

Last I heard, all Ryzen CPUs were overclockable as long as you use Ryzen Master: https://www.amd.com/en/technologies/ryzen-master

AMD explicitly says all their CPUs are multiplier unlocked.

 

Or flawed firmware? The best ssd speed can't overcome bugs or flaws http://forum.notebookreview.com/threads/quick-question-about-windows-10.834104/page-2#post-11046721

Maybe you have an idea how to fix this mess?

 

Could be... but then it boils down to the following:
Whose firmware?
AMD's AGESA, OEM's implementation of it or the SSD firmware?

 

Would be pretty weird if it was the SSD firmware. The tested ssd's ain't exactly new released models. Regardless, not a nice experience for the affected owners. If they I mean the OEMs went with AMD I expect their engineers got guidelines how to implement correct functionality in firmware. Even a nearly blind man can see that +2min boot time is totally wacko. Even for old spinning HD.

 

It is wacko indeed... but AMD may not be specifically to blame here .

Take Acer for example and their BIOS for my PH517-61 (Ryzen 2700 and Vega 56).
When all 4 RAM slots are used, RAM speed drops to 2100MhZ or 2400MhZ regardless if all RAM sticks JEDEC's are rated for say 2933MhZ and lower timings... on desktop though, such problems would seldom occur (and if they do, OEM's usually release BIOS updates to remedy that with say integration of XMP profiles - whereas laptops with Zen CPU's don't have XMP profiles available in the BIOS at all), while on desktops you can easily use even faster (3200-3600MhZ) RAM with Ryzen 2700 and 2700x for example (which of course greatly improves performance in latency dependent software such as professional programs and games).
My Acer does take a bit to POST (nowhere near 2 minutes of course... but I would say it takes about 5 to 10 seconds to POST and then another 10-15 seconds or so to reach Windows).
Could it be better?
Of course.
Will Acer release BIOS updates to remedy that? Very unlikely because they stopped supporting the unit about 6-8 months after releasing it (oh and, Acer never updated their GPU drivers for this unit, they still have outdated ones they released the first time around- fortunately, AMD's desktop drivers do work easily with my laptop so no issues there).

OEM's are also responsible for using low ROM chip sizes on AMD mobos which in some cases prevent integration of latest AGESA and support for Zen 3 (initially there was also concern about supporting Zen 2).
They tend to fill up ROM chips with flashy UI for UEFI - and when the flashy UI was reduced to something more basic, they suddenly had more than enough space.

I would imagine that there's a possibility that AMD's own firmware might be flawed in regards to how long it takes to POST and subsequent booting times being affected by OC CPU's, etc., however, I think its very likely that this has more to do with OEM's implementation than with AMD themselves.
I'm not saying AMD cannot do wrong here and I'm sure there is plenty they can do to improve in the firmware... but this wouldn't be the first time OEM's have dropped the ball on AMD.
Some OEM implementations of AMD gpu's were abysmal where they caused malfunctions such as black screens and BSOD's... Asus Navi GPU's for example come to mind, and it was later discovered that they apparently did no testing to determine the pressures of the cooling design to ensure it was actually working... they DID try blaming AMD until they realized that the burden of testing GPU's which they created to make sure everything works was actually theirs to bear (not AMD's).

 

I guess I don't have the boot issues with my G14, it typically takes about 7-8 seconds to boot up from a shutdown.

 

As expected, Tiger Lake is what Intel said it was.

See:
https://www.anandtech.com/show/16084/intel-tiger-lake-review-deep-dive-core-11th-gen

AMD is still a lessor second-place choice in the mobile computing space for productivity-based workflows that a mobile form-factor is best suited to (and it has never been first). For any workloads short of what a mobile workstation-class platform can provide, Tiger Lake is the best choice for most users once again.

Intel once again provides the best application of tech to solve actual problems (i.e. real-world workloads/workflows). This is vision. This is follow-through, this and proven time-tested compatibility and reliability. AMD's answer to Tiger Lake had better be bold, loud, and clear (and not have to do with more cores either). This means beating Intel in battery runtime efficiency for most users workloads, offering the same experience when on battery power as on AC power, paying attention to details (like booting up, but so much more too).

I have little doubt that AMD will come hard and heavy with big HP and many 'technical' wins, but little to show for when in the hands of consumers like me who want something obviously better than what they had before.

What Intel needs now is manufacturers with the vision to see 'Evo' through (I want to see options with the 28W specs offered/advertised and with Adaptix enabled/switchable for all power levels. Yeah, Intel, these should be requirements and not just optional for the Athena 3 platform, next.

See:
https://arstechnica.com/gadgets/2020/09/hands-on-with-intels-i7-1185g7-tiger-lake-prototype-laptop/

 

I looked at the results from the whole review and I'm not sure how you can say Intel is better for MOST users. It looks like they trade blows in numerous areas. If you want to run SPEC2006 or do lots of java you should get an Intel (not sure what this actually proves, lol). If you want to render videos then AMD is better. Gaming is a mix between the two, I'm glad Intel finally stepped up their iGPU performance. All and all I'm not impressed with their launch.

 

Is this a souped up spec? I hv no idea...
But, as a fair comparison for CPU performance, why can't Intel send out evaluation samples that with RAMs and storage type matches to those of other makes?
Dave from D2D mentioned that he was told not to do any benchmarking, while Linus gets to do it?!!

Fair game, some might say. But to me, a fair game is when competitors are required to compete in their gears with close specs, yet their engines can be little different...

 

Most users will enjoy a more responsive/snappy platform (highest single core performance by a wide margin on mobile). Most users will appreciate up to 30% greater battery life. Most users will be happy to have this wrapped up in an Evo based platform. Most people do not render video and even if they do, they won't be depending on a 15W/28W mobile platform to do it on. Yeah, gaming depends on the games the user is interested in, but if they want any of the above, then it is easy to 'settle' for a bit less on the gaming side if any of the above is really important to them (and it should be if snappiness and portability is important too). If you are not impressed by what Intel has claimed and was confirmed by third parties so far, then you're looking at the wrong segment to satisfy your computing needs.

There is a disconnect between 'fair comparison of CPU performance' and comparing a brand-new platform with new capabilities, to me. I don't care what IPC has increased (if any). I don't care if a memory subsystem is faster or slower. The bottom line is does it perform better overall. That answer is 'YES'. A lot better. Limiting the newer platform to the old/obsolete tech going out is not just unfair for either, but it isn't fair to the consumer too. Consumers are not going to get a platform with IPC differences, they're going to get a better (or worse) platform overall, period.

We're not comparing two (or more) drivers here on a specific racetrack. We're comparing how far (productive) a single/identical (via real-world workloads/workflows) 'driver' gets when in one 'car' (platform) vs. the other 'car' (platform). Of course, it isn't fair to the old tech. At least, it shouldn't be if we're really advancing and not just playing a marketing/numbers ('score') game.

The details are available for what you seek ( yawn). But those details mean nothing if they don't translate into a tangible improvement over what was the highest example of the specific form factor that we're comparing it to now.

Mobile computing isn't about a sole product having the finest possible hardware/tech inside. It is about having the right mix of hardware and software/firmware at a price-point that will sell.

A fitting example is the Surface Duo. While sounding ridiculously expensive at first, when compared to something like even the Samsung Note 20 Ultra (at the same price) it becomes apparent that the 'ridiculously expensive' piece of kit is the Samsung and not the Duo. Why? Even with lower end specs on 'everything' important (CPU, camera(s), etc.), you can do things with the Duo that are impossible on the Note 20U. Productivity blossoms on the Duo in ways that are impossible on the Note (even with their superior, (for now), s-pen).