Intel Broadwell Thread

[Shivansps]

By the way, Jarred fixed it again to say its not Core M.

"The tablets will be using Bay Trail processor, Z3735F (quad-core up to 1.83GHz with dual-channel RAM."

"Text has been updated today. Dell emailed me to apologize and let me know they sent out the wrong specs; the Venue Pro 10 will be Bay Trail, which fits the pricing model a lot better. Too bad...."

If they wanted it for $330, it probably would have needed contra-revenue, but rather than being $20, it probably requires $50.

Z3735F and D are single channel only, and max 2gb too.

 

[dahorns]

Curious that Ncheck doesnt display the FPS, must be something like 0.5 FPS when comparing to what a Beema had rendered after 47min of Furmark...

I may be misunderstanding you, but it says FPS=5 at the top of the picture. That was the current FPS at the time the screen capture was taken. You can see the same thing in your Beema screen capture--the current FPS was 3.

 

[Abwx]

I may be misunderstanding you, but it says FPS=5 at the top of the picture. That was the current FPS at the time the screen capture was taken. You can see the same thing in your Beema screen capture--the current FPS was 3.

Still the look of the fur doesnt correlate with 1h06 min rendering time, how could the Beema render a more achieved pic just after 47mn and at only 3 FPS average..?.

I dont think that Notebookcheck did accidentaly display only a part of the screen, why removing min/max, averages, cores and GPU clocks if they really wanted to prove that their sayings are accurate.?.

Since the laptop was sent by DELL i smell some deliberatly friendly review, not the first time that this site is doing so, after all it s a business and they certainly have not the means to buy even only 10% of the items they are "rewiewing"..

 

[jj109]

Still the look of the fur doesnt correlate with 1h06 min rendering time, how could the Beema render a more achieved pic just after 47mn and at only 3 FPS average..?.

I dont think that Notebookcheck did accidentaly display only a part of the screen, why removing min/max, averages, cores and GPU clocks if they really wanted to prove that their sayings are accurate.?.

Since the laptop was sent by DELL i smell some deliberatly friendly review, not the first time that this site is doing so, after all it s a business and they certainly have not the means to buy even only 10% of the items they are "rewiewing"..

There's a simpler explanation:

You're grasping at straws.

 

[dahorns]

Still the look of the fur doesnt correlate with 1h06 min rendering time, how could the Beema render a more achieved pic just after 47mn and at only 3 FPS average..?.

Does the "fur" become more defined in Furmark? I thought it just rotated the furry object in front of the background.

 

[Abwx]

There's a simpler explanation:

You're grasping at straws.

What is your argument, technicaly speaking..?

None, you would be hard pressed if asked where s the straw, actualy you could find one, that is in your post...

Does the "fur" become more defined in Furmark? I thought it just rotated the furry object in front of the background.

Furmark is a rendering of a fur, hence the name, it s not simply loading the GPU with power consuming functions, indeed the rendering yield a score.

FurMark est benchmark OpenGL qui utilise les algorithmes de rendu de fourrure (fur en anglais) pour mesurer les performances de la carte graphique. Le rendu de fourrure est tout particulièrement adapté pour faire surchauffer le GPU et c'est pourquoi FurMark est aussi un parfait outil de test de stabilité (aussi appelé GPU burner) pour la carte graphique.

"Furmark is an OpenGL benchmark wich use fur rendering algorithms to measure the performances of a graphic card.The fur rendering is especialy adequate to overheat the GPU and that s why Furmark is also a perfect tool for stability test (also called GPU burner) of the graphic card".

http://www.ozone3d.net/benchmarks/fur/index.php?lang=1

 

[Enigmoid]

I would say its mainly P95....when looking at the pic, what happened, or rather what did not happen to the fur despite roughly one hour of rendering?
Curious that Ncheck doesnt display the FPS, must be something like 0.5 FPS when comparing to what a Beema had rendered after 47min of Furmark...

The donut is rotating. On BW-U it is almost completely edge on. Beema shows it face on.

 

[Abwx]

The donut is rotating. On BW-U it is almost completely edge on. Beema shows it face on.

Still it s rendered, as for the pictures you have displayed the image of the Beema test, i repost BDW pic so you can see if the image is rotated but also identicaly rendered, i say that its not as well rendered, what do you think of this pic ? :

Also when i read reviews i d like to have numbers posted, not numbers in one review and no number in another one, i wont take the reviewers words at face value, indeed if they were telling the truth why truncate the pic.?.

I say that if their sayings were accurate they would had proved it with a full screen pic, if there was nothing to hide why not post the thing..?..

 

[dahorns]

Still it s rendered, as for the pictures you have displayed the image of the Beema test, i repost BDW pic so you can see if the image is rotated but also identicaly rendered, i say that its not as well rendered, what do you think of this pic ?

It renders the donut rotating in front of the background. It doesn't change the image quality. Compare: https://www.youtube.com/watch?v=5HsTowN2fZk

Also when i read reviews i d like to have numbers posted, not numbers in one review and no number in another one, i wont take the reviewers words at face value, indeed if they were telling the truth why truncate the pic.?.

The information you're looking for (fps) isn't really the point of the stress-test. They're just trying to show you what the clock speeds are, and the screen shot does that just fine.

At any rate, the current FPS is indeed displayed on both screen shots, so I'm not really sure what you're complaining about.

 

[coercitiv]

For reference, an undervolted Haswell Ult requires ~21.5W to mantain 2.6Ghz in Prime 95. Dropping turbo boost max power to ~19.5W will limit clocks to around 2.5Ghz (2.4-2.6). At 15W TDP it will stay at 2Ghz.

Even if we were to consider BW to be 30% more efficient than HW, it would still require ~15W to maintain 2.5Ghz with Prime 95 alone. Add Furmark in the equation and it should easily go over 18W, unless the new gen GPU only requires 3W. Again, for reference, the undervolted HW can saturate a 19.5W power limit running only Furmark (CPU Package 19.3W, GT Cores Power ~13.5W).

 

[Essence_of_War]

Anandtech just dropped their XPS 13 review w/ Broadwell-U.

http://www.anandtech.com/print/8983/dell-xps-13-review

Battery life is VERY impressive.

 

[Sweepr]

Some interesting bits:

With the Core i5-5200U in both of the XPS 13s that we received, we have 24 execution units, compared to only 20 on Haswell-U. In addition, the 14nm process should help with throttling. The FHD model (1920x1080) arrived with a single 4GB memory module and the QHD+ version came with 2x4GB, which gives us the chance to check the performance differences between the single-channel memory and dual-channel memory.

Multithreaded is where the new 14nm process can really shine. The lower power usage means that there is more headroom for the CPU to keep the clock speeds higher when all cores are active, and it shows a great result here compared to the Haswell-U series parts. x264 HD also shows favorable results for Broadwell, with the i5-5200U outperforming the old i7-4500U. The Dell XPS 15, with the full quad-core, eight-thread CPU, unsurprisingly outperforms all of the U series CPUs.

As Essence of War said, incredible battery life:

Broadwell-U is part of that, but the display is also a big part of that score, especially on the light battery life test, and Dell would have had to work on the entire package to achieve this kind of battery life result. Any one choice that was a power hog would have dragged the scores down.

Easily one of the (if not the) best Macbook competitor from the Windows camp today.

 

[witeken]

Battery life is VERY impressive.

Yes: IGZO is a vast improvement.

And it is obviously much superior than Apple's offering with Haswell and a cheap low-res TN panel.

 

[III-V]

Amazing what a decent site can do for a product. I remember reading notebookcheck's review, and wasn't nearly as impressed. I went back and checked their numbers, and sure enough, they're pretty similar.

The presentation of data is just so much better at AnandTech... notebookcheck is so cluttered, and you often have to click links to get to the important charts to show.

 

[kimmel]

The best part of the review is that we can look pretty much side by side how big of a difference the panel+touch makes on battery life.

 

[asendra]

I don't think anyone doubts IGZO displays help a lot with battery, not even Apple.
But is one thing to choose a panel that is going to be used in 300k devices at best each quarter, and another matter entirely to choose a panel which needs to be in ~2,5-3m devices.

It will be interesting to see how it fares the new Macbook Air against it. I'm betting that will use broadwell also with a retina display. (I'm hoping the rMbp get skylake around september, or earlier! though)

 

[dahorns]

I don't think anyone doubts IGZO displays help a lot with battery, not even Apple.
But is one thing to choose a panel that is going to be used in 300k devices at best each quarter, and another matter entirely to choose a panel which needs to be in ~2,5-3m devices.

Independent of other factors, why should the # of devices matter?

 

[witeken]

I don't think anyone doubts IGZO displays help a lot with battery, not even Apple.
But is one thing to choose a panel that is going to be used in 300k devices at best each quarter, and another matter entirely to choose a panel which needs to be in ~2,5-3m devices.

IGZO is used in the iPad Air. I guess it's a fairly mature technology by now.

 

[asendra]

Independent of other factors, why should the # of devices matter?

I'm not sure what you mean by saying "Independent of other factors".

But what I meant to say is new technologies take time to ramp up production, preparing new processes, adapting factories, achieve good enough yields, etc
Apple normally works with larger scales, as is the case here (Dell XPS 13 vs Macbook Air) so I'm guessing they can't jump as soon as other companies to try new technologies in some of their product lines.

 

[III-V]

I don't think anyone doubts IGZO displays help a lot with battery, not even Apple.
But is one thing to choose a panel that is going to be used in 300k devices at best each quarter, and another matter entirely to choose a panel which needs to be in ~2,5-3m devices.

It will be interesting to see how it fares the new Macbook Air against it. I'm betting that will use broadwell also with a retina display. (I'm hoping the rMbp get skylake around september, or earlier! though)

IGZO has probably gotten pretty cheap at this point (relative to initial costs).

 

[dahorns]

I'm not sure what you mean by saying "Independent of other factors".

But what I meant to say is new technologies take time to ramp up production, preparing new processes, adapting factories, achieve good enough yields, etc
Apple normally works with larger scales, as is the case here (Dell XPS 13 vs Macbook Air) so I'm guessing they can't jump as soon as other companies to try new technologies in some of their product lines.

Ah, so you're just worried about the supplier keeping up with production. Fair enough.

 

[Qwertilot]

Although Apple do also quite often use their quite outlandish revenues/cash pile to make quite sure they can get enough of genuinely nice things

Maybe as likely that it just wasn't ready for their update last year.

 

[Qwertilot]

On another note, are (which of?) the gaming figure comparisons like for like comparisons with Haswell? It looks like it might be non trivially better in actual games from the graphs, more so than the raw figures, but hard to be sure.

Won't be interesting for that until we hit iris pro of course and Broadwell K is also due to get a much higher power budget which should help a chunk more.

 

[witeken]

For the lack of a better thread,

Here's a great article, interview in fact, for one's reading pleasure. Topics include R&D, EUV, 450mm, 7nm.

http://semiengineering.com/one-on-one-dave-hemker/

 

[DrMrLordX]

Won't be interesting for that until we hit iris pro of course and Broadwell K is also due to get a much higher power budget which should help a chunk more.

Correct me if I'm wrong, but hasn't Intel canned all Broadwell desktop CPUs (including Broadwell-k) except for a 65W unlocked Broadwell Iris Pro part? Which, I guess is going to be Iris Pro and Broadwell-K at the same time?

Or am I reading the roadmaps wrong?