HP Spectre 13T-3000

Ironically - I was in the same boat about the coupon. In fact, I bought my Envy 15 three years ago because of a 20% coupon. I received my 13t exactly 3 weeks ago, and immediately got on the phone with HP customer service today after the coupon came back into effect. The rep initially told me the coupon was entirely used up, which I let him know that the coupon was actually in effect again. After checking, he said that I would have to return the laptop, start a new order just for the coupon to apply. He mentioned that this was because the laptop was ordered in October, and the coupon was only good in December. The difference is pretty substantial - over $200 additional savings for the same configuration (i5, 8 gigs or ram, 256GB hard drive, upgraded wifi, high-res 2560x1440 screen).

Now, while I understand the technicality behind this - you would think the supervisor would have realized they would have to pay shipping twice (return and new one) and take back a used laptop would be a lot more wasteful and unnecessary then just applying the coupon to the existing order and call it a day. But the supervisor rejected this argument (who knows why - i didnt get to actually talk to him) and so I started the process of creating a new order and returning the one I had just received...left me a little surprised that such waste would happen.

I am happy with the laptop overall, a bit underwhelmed with the HD 4400 integrated graphics card (tried to run Dungeon Hunters 4 - free download from the Windows Store), but its perfect for browsing, office related spreadsheets and music/videos. battery life is great so far, and the backlit keyboard does not bother me in the least. The active cooling does get a bit strange, and i swear I can also hear a slight high-pitched whine, but i personally think my ears tend to pick up these whines in all laptops (ie. battery plugged in to my tablets or phone and i sometimes hear a high-pitched whine - its strange really...).

 

Hi All,

This seems to be the first review available for HP Spectre 13.

HP Spectre 13t Review - eCoustics.com


Regards,
Kal

 

An alternate link to the review:

HP Spectre 13t review | Digital Trends Reviews

 

Btw... I ordered the Smart Buy model (Core i5, 8gb ram, 256gb SSD, QHD screen & AC WiFi) for $1135 (inclusive taxes) on Nov 29 and with an estimated ship date of 17th Dec. But my laptop has been shipped today with estimated delivery date around 16 Dec.

Hope this information helps for folks who are awaiting their's to be shipped.

 

Thanks for posting that review. Nice to finally see a battery life benchmark.

Can someone try running Cinebench R11.5 while on AC power? It's strange that they got lower CPU benchmark scores.

 

I ran SiSoftware Sandra Lite, but I ran the 2013 version and got a score of 4.2Ksomethings.

I installed the 2014 version and thought I ran last night, but apparently it never actually launched. I will try again tonight.

One thing I noticed is that the laptop does get quite loud when it runs things like benchmarks. This isn't too surprising when you think about it: if you put any amount of computing power into a small enough space you will need to blast a lot of air to keep things cool under high load. Either that or you want to try and use the whole metal body as a heat sink in some clever way (maybe HP does both).

I noticed that the body does get warm under high load, I was wondering if one could attach a liquid cooler to the bottom and increase the cooling capacity that way (I think I've seen some laptop coolers that effectively do this).

Anyway, I still think this machine is great. Over all very good tradeoffs were made in the design. You just can't have everything. In the future we'll all have 2^32 core 3.2 TeraHZ hearing aids with ZHD synaptic interfaces and TeraBit WiFi (or 4kG) that run for 10 years before you throw them away. They would replace your phone, laptop, desktop and TV. =P

I wish I hadn't managed to accidentally install so much mal/ad-ware on my computer while trying to get SiSoft Sandra on this thing. Doh!

 

Fred, take some comfort knowing that your sacrifice to give all on this thread more benchmark data is recognized and appreciated. If you want to post your PayPal account, I, for one, pledge to contribute $5 to help you purchase paid versions of benchmark software to reduce the ads and crapware you are dealing with for the "greater good!"

On the topic of that review's finding the Spectre 13T low on performance, have any of the owners here sensed this - in your gut - either before or after reading the review?

Sent from my GT-N5110 using Tapatalk

 

I have a private performance benchmark that I have used for about 75 different PCs over the last 20 years. My i5-4200u Haswell performs about the same as my former i3-3217u Ivy Bridge.

Good luck, Jim

 

I thought the main advantage to the new Haswell processors was that they are lower power, not necessarily higher performance.

Still, I would be surprised if the 13t-3000 were radically different than the other contenders with the same Haswell configuration.

For my part, I have not found the laptop to be lacking in performance in any unexpected way. I won't be doing much data crunching on this thing just the occasional old game.

I'll see what I can do about running more bench marks. This is also a great way to test the laptop while I'm still well in the return period.

 

I thought the lower power came with a performance hit. I saw a discussion on another forum where people were complaining about lack of full voltage "M" processors on the Haswell series PCs. The sentiment there was ":give me full voltage, high power processors and the heck with battery life".

Regards, Jim

 

I'm very surprised by that result. The i5 4200u benchmarks about 50% faster according to most sites.

Intel Core i5 4200U Notebook Processor - NotebookCheck.net Tech
Intel Core i3 3217U Notebook Processor - NotebookCheck.net Tech

PassMark - Intel Core i5-4200U @ 1.60GHz - Price performance comparison
PassMark - Intel Core i3-3217U @ 1.80GHz - Price performance comparison

 

I'm well out of my knowledge zone here, so I'll point to Wikipedia:

Haswell (microarchitecture) - Wikipedia, the free encyclopedia

Haswell is a new architecture with some modest performance improvements over ivy bridge, but the main advantage is the lower power consumption. The bigger deal is the integrated graphics which is supposed to be much better than the previous generation.

 

Yeah, comparing apples to apples (entry level i5 ULV Ivy Bridge vs Haswell, for example) you should see a slight performance bump.

PassMark - Intel Core i5-3317U @ 1.70GHz - Price performance comparison
PassMark - Intel Core i5-4200U @ 1.60GHz - Price performance comparison

Individual laptops could have design flaws that cause thermal throttling, however, which may be what's going on here.

Edit: And if it is related to thermal throttling, that typically does not have nearly the effect on real-world usage as it does on benchmarks.

 

that is weird that the i5-4200u is slower than other i5-4200u cpu's

I wonder why that is

 

Yeah, in my normal use, I would rate the i5 Haswell slightly faster than the the i3 Ivy Bridge. My performance test may not measure CPU crunch power properly. All it does is calculate PI to 11 decimal places starting with division of two small integers. All crunch power, no graphics, no write to screen, etc in the timed segment. And CPU speed is used in the final calculation. And who knows what that is for certain with Turbo Boost and CPU throttling?

Regards, Jim

 

The "U" processors since they came out has always been about minimum power consumption over performance. This is not to say that the "U" processors are dogs, because they are not, they were/are intended mainly for the Ultrabooks market. And who are the Ultrabook Users: People who demand long battery life, who primarily does emailing, surfing, lite gaming, video watching and lite video/photo editing, who understand that performance may not be that of the "M" processor, all wrapped up in a Ultra portable case. I believe that the HP 13T has masterfully fit that bill, especially when fitted with the i7-4500U.

The problem comes in when users get their Ultrabook and it doesn't perform as they think it should because they compare it to the benchmarks that are out there to other vendors. Or user sometimes do more then what the system was design to do. (I use to be very guilty of this and frustrate the heck out of myself).

I agree that if Vendor A designs an Ultrabook with those exact same components as Vendor B, that both vendor's Ultrabook should theoretically result in similar performance. However, what differentiate vendor A Ultrabook is that it had a better design (Not speaking of cosmetics) and perhaps better thermals, tweaked drivers, etc. The point of all of this is this: If I get wrapped up into why my Ultrabook is not performing as like the next Ultrabook (especially only by a few points) I will drive myself crazy.

The review that was posted was a great review, and of course when he got to the performance aspect we all barked, including myself ("How dare he") because we are all admirers of the HP 13T. And most of us who seen this review has already commented and replied by saying "Well I don't see a performance Hit with my 13T". And that's my point, the question that you have to ask yourself is your HP 13T Spectre performing satisfactory to you? Is it fulfilling my intended needs and my needs didn't creep or morph into something else (more demanding needs.

I think the world of benchmarks because it let's me know where a particularly component or system should be performing at, especially when I am building a particular component/system for a particular purpose. Back in my overclocking days, I was one of the 1st followers of Tom's Hardware and AnandTech (Before they went commercial) and I use to overclock my AMD Athlon Thunderbird or Intel Pentium Pro (Yes, Pentium Pro, I am dating myself) only to find myself frustrated because my neighbor's self-built system out performed mine because he brought a better fan and better thermal paste. ...I digress. Here's the gist, Benchmarks should only be used as a reference point.

Don’t get me wrong, if the HP Spectre 13T is lacking severely in performance behind other vendors with similar components, then let’s investigate and find out why. But I simply don’t think that is the case with the 13T. When I had mine, albeit with an i7-4500U, I didn’t see any performance hit. And I tried to thrash it, (multiple windows with video running, opening up large Visio/Excel Files, while watching the OS performance monitor etc.) it performed respectfully.

Lastly, I would just like to say, when it comes to Ultrabook, buy the best processor that you can afford, and most memory that you can afford, especially if a vendor is having a sale. People say that there not much of a difference between the i5-4200U and the i7-4500U and I would agree to a certain point. But to me the big difference between the two that matters to me and that often get over looked is the cache. Yes, I know it only 1MB difference, but you factor that in with the processor’s min and max speeds, it helps when you find yourself doing multiple tasks. How many of you could remember when the CPU came with 256K or 512K cache or when Intel made that stupid mistake of going cacheless with the Celeron

Don’t put too much into the benchmarks. Have fun and enjoy your new HP 13T Ultrabbook as it is one of the best that are out there.

BTW, Those of you who know me from the beginning of this post, yes I still have my wonderful Alienware, but I also just re-purchased the HP 13T Specte a 1 1/2 week ago (still maxed out, minus the Windows Pro) and Believe it or not it already shipped. With the 20% off it was simply a no-brainier; it the best value there is on an Ultrabook.

Thanks for reading my Novel :hi2:

 

The processor is the same speed, the issue is that benchmarks tend to push a processor so that heat becomes an issue. With their ultraportables HP has historically tried to throttle the CPU rather than spin the fan up too loudly. This leads to benchmarks being lower but won't typically affect your real world performance. There are some exceptions though, like my HP dm1 starts to throttle in some games at only ~75C (Tjunction for the CPU is 100C) leading to some games being unplayable.

 

I agree. I don't concern myself with benchmarks. I guess I'm just not knowledgeable enough to care.

I just try to decide if my current laptop does everything that want/need it to do at acceptable speeds. If it does, great, I'm happy. I don't concern myself if another laptop is a millisecond faster.

It like fast cars, there is always someone out there with a faster car.

You'll never enjoy what you have, if you are always pursuing the latest and greatest.

I think that the Spectre 13t-3000 is one heck of a deal, especially with the 20% off coupon.

 

I've been playing around with the ultrabook to see if I candetect any problems before Christmas and the return deadline. I have encountered something wierd and wondering if anyone else had this experience. Or if not can comment on what is possibly going on and if I have a defective model.

I've been primarily internet surfing on the Chrome browser using a combination of touchpad/keyboard and touchscreen input to navigate/use the websites. After being on a website for quite a while ~20 min, the touch screen input capability either stops working totally or doesn't work as it has previously or most effectively. The curser/touchpad/keyboard can still be used for navigation but when I try to use the navigation button via touchscreen it does not register or work. So far this has occurred on 2 different websites for me, Victoria's Secret and Amazon. And I've been able to replicate the phenomena on both websites. What happens is when I select a navigation button via touchscreen it won't work, even though the same button on a different screen previously worked 20+ times (ie Next page selection). The other touchscreen navigation options on the page also will not work. I am able to navigate by selecting these buttons with the touchpad and enter key. The other thing that happens is after a while touchscreen page scroll stops working properly. It reaches a point down the page when further swipes will not move the screen image any more down the page and it just returns to the same point. And I've also had the page scroll function on the touch screen just stop working as well. Again, when I relaunch the browser the touch screen features initially work fine, it's only after being on a website for some time.

Anyone else have this experience. Does anyone have an opinion as to whether it is the touchscreen itself or a software or website server issue? I'd appreciate any input you can provide. I'm officially worried now.

 

Does it also happen using IE?

 

Sounds like the i5-4200U's turbo boost is not kicking in. Looking at a review of last year's Spectre XT 13-2000, it looks like the Spectre does disable turbo boost at lower temperatures than most other laptops.
For typical usage this isn't a problem. Dual-core Haswell is very capable at 1.6ghz, and gaming shouldn't be affected since the ULV CPU's disable turbo when stressing the GPU anyway, but this is pretty unfortunate for those who want to stress the CPU for compiling code, rendering videos, etc.

Can someone check if there's a BIOS setting to disable this sort of thing?

 

Just thought I'd confirm that I got about the same performance number they mention in the review. I actually have the i7 so you'd think it would be a bit better: 26.26GOPS

I'll check the bios when I get the chance...

 

Well, I checked the BIOS but I didn't see anything too promising. You can disable 2 intel related things: Intel Smart Connect and Intel Rapid Start. Both are enabled by default but can be disabled.

For those of you who need function keys there is hope. You can change the settings so that the "Fn" key must be pressed for the function keys to be action keys making them F1-F12 by default. In other words, to brighten the display you would have to hit the "Fn" key and "F3" at the same time, but you wouldn't need to hit "Fn" if you just wanted "F3".

The HP CPU nerfing is a bit of a mystery. I wonder if they couldn't keep the system cool enough and had to step something down. Maybe they'll unlock the BIOS so we can tool around with it some day.

 

One other bit of info - I ran the Intel Processor Identification Utility and apparently the bus on this system is running at 49MHz.

The utility also states that the processor was manufactured to operate at 100MHz. I don't know if this gets stepped up and down, but this would definitely explain the performance numbers.

It would be great to see some comparisons with similar hardware (the similarly configured Sony, Asus, Samsung, Acer and Dell ultrabooks). They had other issues though. I'm not sure faster bus speed would compensate for them...

 

Sorry to flood the forum, but one more word on this: I also looked at GeekBench:

Hewlett-Packard HP Spectre 13 Ultrabook Benchmark Chart - Geekbench Browser

The lower datapoint is mine, but the other two entries give much better numbers for this laptop, so I'm almost certain some kind of throttling is going on and I bet it can be disabled somehow. It may require uninstalling some software or some such thing.

EDIT: and I should add that the other two entries are very close to the competition listed on this site (Aspire S7-392 and Dell XPS 12)

 

Mm, that may just be a bug with Intel Processor Identification Utility. It would be very unusual for the i5-4200U or any Haswell CPU to be running with a 49Mh bus speed. Typically it's the multiplier that changes, while the bus speed stays constant (frequency = bus speed * multiplier).

Can you try out CPU-Z and see what it reports for the bus speed? If you could track the frequency/core speed while running a CPU benchmark, that would be great too. I'm wondering if it turbo's at all when stressed. Without turbo, your i7-4500U should hit a maximum 1.7ghz, but with turbo, you should see 2.7~3.0ghz. This would account for the very large benchmark differences. The i5-4200U should reach 2.3~2.6ghz with turbo.

I have a few thoughts on attempting to enable turbo... I'll post them in a moment. I really hope this can be resolved.

 

Well I take some heart in knowing that some others out there have resolved this, namely the other two people that uploaded there geekbench results

I'll let you know what I find out.

 

Hey Fred just wanted to say thanks for digging around and trying to find out what's going on. Receiving mine in a few weeks and would love to know if there is some solution. Really is quite a strange occurrence.

 

Ah, I didn't realize I was looking at only Spectre scores, thought the other two were the S7 and XPS 12 haha.
One difference between your benchmark and theirs is that you're running a newer BIOS version (.07 v .06). There appears to be an even newer version available, v.09. Maybe try upgrading to that? The release notes don't mention anything CPU related however.

Anyway, here's a couple of things to try to enable turbo:
1. Ensure that the maximum processor state is at 100%. By default, the "High Performance" power plan should have this. You can check this value by right clicking the battery icon->power options->Change plan settings->Change advanced power settings-> and finally this window.

2. See if any options are unlocked in Intel Extreme Tuning Utility. All of the options are likely to be locked down in a laptop, but the Lenovo Yoga 2 Pro Ultrabook actually had a number of the sliders unlocked, so might as well check.

 

I was doing some more reading, and this review on the Spectre 13 x2 mentions the manufacturer artificially limiting the power consumption of the CPU from 11.5 W to 7-9 W, for the sake of temperature management. This is with the native i5-4202Y, but it seems that they may have hamstrung the ultrabook version of the Spectre 13 in the same way. I'm guessing there may be overheating issues otherwise? Pretty sneaky stuff.

 

Looks like it still hit the normal turbo frequencies, so I'm not too sure if that's what we're seeing on the Spectre 13t. From the other review, I don't think there was any unusual throttling when stressing both the GPU and CPU. It just seems like turbo is prematurely disabled to keep CPU temps as low as possible.

 

Ok never mind then! Good to know.

 

Does this mean there is no reason to upgrade to i7?!? This would be worse than a bait and switch since you paid for the higher level of performance... as did I when I just placed my order.

 

I see something similar, The touch pad stops scrolling on some web sites. But, a small scroll using touch screen or arrow key gets it going again.

Jim

 

Okay, here are some screen shots from CPU-Z. I realize some of these tabs are pretty useless, but I thought I'd include the batch:

 

The jury's still out on this. I would have to measure frame rate play a game or some such thing.

 

Mine just shipped (i5, Hi res, 256G) so I'm a relatively happy camper - it was scheduled to ship on the 17th, ordered on the 1st.

I'm not going to get too torqued up about the processor speed. I'm not planning on encoding a 2 hour movie or compiling a gajillion lines of code on this machine - I need something very lightweight, long battery life, and enough horsepower to do some moderate to heavy MS Office stuff. I think this machine will fit that bill just fine - and the extra bling don't hurt any either

Also just ordered this sleeve. I was looking for something that will keep down the scratches that might occur while it's in my regular bag and this looks like a good fit. I'm a little concerned about the magnetic closure, though. I'll report back when all is in my eager little hands.

 

Excellent advice, I will check these things out. I have no problem supporting max performance while the unit is plugged in.

EDIT: I might hold off on the BIOS update. I prefer not to mess with these unless there's a real problem.

 

I have a private performance benchmark that I have used for about 75 different PCs over the last 20 years. My i5-4200u Haswell performs about the same as my former i3-3217u Ivy Bridge.

Good luck, Jim

I guess we all read the same stuff and come away with slightly differ impressions. My takeaway was that Haswell ULT cpus performed at very close to the same level as their Ivy Bridge equivalents and that it is the "Y" designation that indicated even lower voltage than "U" such that one should expect the Haswell U cpus to perform like their IB counterparts (though better in graphics even with the 4400) and the Hasw. Y would perform, as in Jim's test of Spectre 13t with the i5-4200u, about like one class lower in IB lineup. So, the incongruity seems to be that the i5-4200u should perform about like the i5-3317u rather than the i3-3217.

This leaves me with two questions:

1) What is the % difference in performance between the iB i3 and the IB i5 (the latter being what the 4200u should equal) or, put differently, is this enough to care about? Note that the i3 doesnt have turbo boost yet on Jim's benchmark it still equals the 4200u with turbo. Have you checked to see that your power plan (or using HP "Collsense) isn't inhibiting turbo boost in the test?

2) I would want to know why this is because it surely is not "normal" from the many Haswell i5u laptops tested

***NOTE: I learned from Lisa Gade, the owner of mobiletechreview.com why there are no published reviews of this product. It seems that for several months now HP has pretty much discontinued its review distribution program and that's why no one has reviewed this one - or hardly any HP that was released in the last month or two. Lisa said she doesn't know why they've done this or if it is considered a short or long-term policy. I told Lisa I would buy one, loan her mine to do a review on and keep it if she gets good #s or return it if she doesnt! I dont think she was comfortable doing this, lol.

Sent from my GT-N5110 using Tapatalk

 

I ordered this leather sleeve but was looking at that one as well. Review/pictures would be great.

 

Hey, I forgot to thank your for the very kind offer! I will keep plugging away at benchmarks because I do get a kick out of it!

 

Yes, I think I reached the same conclusion you did w.r.t. the comparison of IB and Haswell processcors.

Since I saw a big drop in performance compared to other benchmarks for this laptop configuration (and other laptops with the same configuration) I think there must be a setting somewhere. I will spend some time on this tonight and look into "Collsense", the power management settings and look at the Intel Extreme Tuning Utility mentioned above. As a last ditch effort I could try updating the BIOS, but I'm pretty reluctant to do that unless I really have a problem (I don't need a paperweight that looks just like an expensive HP laptop...).

That's an interesting strategy on HP's part and I guess it worked. Maybe I would have been more reluctant to buy one of these if there had been critical reviews. Reviewers are not like normal consumers in terms of trying to work out problems with HP, so I guess HP also doesn't get as much out of the deal. They probably get more out of working with customers who have a vested interest in ultimately having a great laptop.

 

just got my 13t. need some help. it looks fantastic but i dont think my resolution is right. the desktop looks beautiful but when i use chrome it definitely doesnt look that nice. if i use the windows 8 chrome it looks how it should FHD, but i want to use the desktop version regular and that definitely isnt looking right.

also just like when i am installing itunes, the itunes box looks a bit pixelated. am i supposed to do something myself to make sure the resolution is crisp for everything or how do i fix this?

 

I don't have a solution for you but I know what you're talking about. I have the same issue with my Lenovo Helix. It must be a Windows 8 thing. Haven't been able to figure out why it doesn't look as sharp in desktop mode.

 

I have some more data abut the tests I performed on the Fujitsu i3-3217u and the HP i5-4200u. As I mentioned before, the program calculates PI to eleven decimal points starting with division of two small numbers. All of the tests except for the last couple were done with 16 bit assembler or some very early tests were with Basic. But, I had to re-write for the current 64 bit systems. The early tests used system time directly, but I was advised that timer measurements were not reliable with Masm32. So, I ended up using CPU cycles as the measurement. And to get better results I did the PI calculation 1,000 times and averaged five runs. But, now with throttling and turbo, getting times becomes a problem. But, since I am doing the calculation 1,000 times I can use a stopwatch to get the time within a few tenths of a second. Anyway here are some numbers that have me spinning.

The Fujitsu i3 takes 34,615,770 CPU cycles to complete the 1,000 calculations. A stopwatch measures 20.3 seconds which translates to 20.3 milliseconds. That would indicate that the 1.8 GHz i3 Fujitsu is running 1,7 GHz.

But, the HP i5 takes 50,885,880 CPU cycles and a stopwatch measures 23.1 milliseconds. And that indicates that the 1.6 GHz HP is running 2.2 GHz. (Turbo)

It just doesn't add up. Why would an I5 Haswell run slightly slower than an i3 Ivy Bridge?

Regards, Jim

 

gah..i thought this comes with windows 8.1?

i have a beautiful 1080p FHD screen and yet everything looks like its crappy 1366x768

itunes also looks grainy, and so does the itunes logo on my taskbar..

is there really no solution to this? this is pretty disappointing. im sure its an easy fix, has anyone else solved this?

 

I agree. I think this is configuration related handled either by the BIOS or by some other software.

 

Well, I will never install anything from CNET again. I got served up a host of cr@pware while installing cpu-z. I just finished uninstalling (hopefully) all of it...

 

There is a BIOS update that addresses the intermittency that you describe.

The new version is 17.0.5.18 and is available on both hp.com and via HP Support Assistant.

Synaptics TouchPad Driver - HP Spectre 13t-3000 CTO Ultrabook | HP® Support

 

Can you use CPU-Z or Task Manager to confirm that the CPU frequency is clocking up to 2.2ghz? From what people have posted here, it sounded like the i5 was stuck at 1.6ghz, the non-turbo frequency.

Yeah, the bus speed is ~100mhz as it should be. Can you take a screenshot of this while running a benchmark? With turbo, you would see up to a 30x multiplier/3.0ghz frequency for a single-threaded load or 2.7ghz for a multi-threaded load. Without turbo, it should be at 1.8ghz.

There should be a ~40% performance improvement with turbo enabled on the i5-4200U. For applications that heavily stress the CPU, this is definitely enough to care about. Turbo frequencies are the most significant difference between the i3 and i5.
Seconding on checking that the power plan is not inhibiting turbo boost. Some default power plans disable turbo boost to preserve battery life.

The strange thing is that those two Geekbench scores imply they had turbo working on the Spectre 13t.

A "fix" would be to disable scaling altogether. This has the downside of everything being much smaller, however.
Otherwise you'll have to wait for those programs to better support scaling.