mboratko

although with this CPU I can now stand to wait 6 months or so and see what AMD Zen 2 brings to the table

it seems like it will work out to be around $1000 for this level of upgrade

I'm now considering going with a dual E5-2667v2 system (3.3GHz base, 4.0GHz turbo) with 128GB DDR3 1866 RAM

@JMY1000 I ended up getting a Xeon E5-1660 v2 (3.7GHz base, 4.0GHz turbo) and it is already performing much much better than the old processor I had.

but, taking into account that the 9900k can easily be overclocked to 5GHz on all cores, and the IPC of the 9900k is already quite a bit higher than the AMD (plus it has AVX512) I'm wondering if I'd be better off spending the extra money after all

Obviously AMD is the more economical option

The main CPUs I was considering was the 2700x or the 9900k

I'm not sure upgrading just the CPU makes sense, the board still uses DDR3 RAM

I'll actually train elsewhere

I want to have a desktop system where I can rapidly prototype models and try things out

I have a 1060 GPU, I'm considering getting a 2080ti once prices come down

I use it as a server and a workstation

Intel Xeon X3450

OK, I switched to my desktop

Given that AMD is coming out with Zen 2 shortly, and it looks like this is going to be a big boost especially for scientific computing, maybe it makes sense to wait then?

I can't recall the model, but it is probably ~ 6 years old and is quad core with HT.

... out of necessity or in order to get comparable performance. I know AMD is worse at AVX2, for example.

It's hard to know without having the processor, all I know is that my current Xeon at 2.8GHz is showing its age, and I want something which is incredibly snappy. I can't foresee all the workloads I may do in the future, but there have been plenty of operations I have run currently where the CPU is pegged at 100% across all cores. As for what I mean by AMD being quirky, I mean (for example) how the 2990wx is actually worse than the 2950x for memory bandwidth sensitive workloads. I'm also wondering if getting an AMD means that I have to often compile libraries instead of using binaries, either..

It's not 5GHz specifically, but it looks to me as though anything from Intel with >= 16 PCIe lanes drops down to 4.2GHz boost clock, and even then it costs $2000. The point is I want a PC which performs as a very responsive development desktop, while also being able to prototype deep learning models. I'd be open to AMD, but I'm not sure CPUs from AMD are advisable for scientific computation, their Threadripper options seem a bit quirky.

Anyone available to discuss a topology question regarding homotopy? I am stuck on an exercise from Hatcher's book, and I would post a question but I really think I need to discuss it briefly...

Is anyone available for a topology chat? I would ask a question, but I really need to have a discussion regarding homotopy for a couple of minutes.

Right, and the other two are the same so there doesn't seem to be any way it could be $-\sqrt 2$

@skullpatrol I traversed the points in a counterclockwise manner. I believe it is a theorem that the direction does not matter in integration with respect to arc length.

@robjohn Great, thanks again for the help!

OK, thanks for the help. This is the first problem on integration with respect to arc length, so I wanted to make sure I understood it correctly before doing many problems incorrectly

So the book's answer is just incorrect then (they had $-\sqrt2$)

Ok, so I get that the total integral is $1+\sqrt 2$

For (0,1) to (0,0) I had $g(t)=(0,1-t)$

@robjohn My book (Apostol's Calculus) defines general line integrals first, and then they go on to say that a line integral of a scalar field $f$ with respect to arc length along $C$ which can be parametrized as $g(t)$ is defined as $$\int_C f \, ds=\int_a^b f(g(t))\lVert g'(t) \rVert dt$$

$\int_C (x+y) ds$ where $C$ is the triangle with vertices $(0,0)$, $(1,0)$, and $(0,1)$ traversed in a counterclockwise direction

@robjohn The problem is: Calculate the line integral with respect to arc length of the following

Sorry :)

Hopefully you get some other responses to your question

Hmm... I guess I can't be of any help then, I'm not sure what the notation would be otherwise

ok

Yes, but Green's theorem doesn't apply here

so how were you applying Green's Theorem?

right

@Paul Can you post the integrand?

the integrand in question looks like a scalar field

Green's Theorem applies to vector fields in $\mathbb R ^2$

(with respect to arc length)

A line integral whose differential is $ds$ is the line integral of a scalar field

In particular, how are you applying Green's theorem if it is a scalar field?

@Paul Is the integrand a vector field or a scalar field?

Again, I'm hesitant to say yes unequivocally because I have not used the notation myself often. That being said, I would think that $dS_1$ would be the same as $ds$ in your typical line integral notation, so yes I believe you are correct.

I believe you can. Compare, for instance, the notation in your book for the divergence theorem and this notation of the divergence theorem. Perhaps someone more familiar with the book will come along with a more definitive answer, however.

What's the source of this notation? I have not seen it written this way.

Ah, yes I have seen it written that way before, I was misinterpreting it.